TikZ and pgf

Manual for version 1.18
\tikzstyle{level 1}=[sibling angle=120]
\tikzstyle{level 2}=[sibling angle=60]
\tikzstyle{level 3}=[sibling angle=30]
\tikzstyle{every node}=[fill]
\tikzstyle{edge from parent}=[snake=expanding waves,segment length=1mm,segment angle=10,draw]
\tikz [grow cyclic,shape=circle,very thick,level distance=13mm,cap=round]
\node {} child [color=\A] foreach \A in {red,green,blue}
{ node {} child [color=\A!50!\B] foreach \B in {red,green,blue}
{ node {} child [color=\A!50!\B!50!\C] foreach \C in {black,gray,white}
{ node {} }
}
};
1
F¨ ur meinen Vater, damit er noch viele sch¨one T
E
X-Graphiken erschaffen kann.
Till
Copyright 2007 by Till Tantau
Permission is granted to copy, distribute and/or modify the documentation under the terms of the gnu Free
Documentation License, Version 1.2 or any later version published by the Free Software Foundation; with
no Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license is included in
the section entitled gnu Free Documentation License.
Permission is granted to copy, distribute and/or modify the code of the package under the terms of the gnu
Public License, Version 2 or any later version published by the Free Software Foundation. A copy of the
license is included in the section entitled gnu Public License.
Permission is also granted to distribute and/or modify both the documentation and the code under the
conditions of the LaTeX Project Public License, either version 1.3 of this license or (at your option) any
later version. A copy of the license is included in the section entitled L
A
T
E
X Project Public License.
2
The TikZ and PGF Packages
Manual for version 1.18
http://sourceforge.net/projects/pgf
Till Tantau

Institut f¨ ur Theoretische Informatik
Universit¨at zu L¨ ubeck
June 12, 2007
Contents
1 Introduction 14
1.1 Structure of the System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.2 Comparison with Other Graphics Packages . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.3 Utilities: Page Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.4 How to Read This Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.5 Authors and Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.6 Getting Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
I Tutorials and Guidelines 17
2 Tutorial: A Picture for Karl’s Students 18
2.1 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2 Setting up the Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2.1 Setting up the Environment in L
A
T
E
X . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2.2 Setting up the Environment in Plain T
E
X . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2.3 Setting up the Environment in ConT
E
Xt . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.3 Straight Path Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.4 Curved Path Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.5 Circle Path Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.6 Rectangle Path Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.7 Grid Path Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.8 Adding a Touch of Style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.9 Drawing Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.10 Arc Path Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.11 Clipping a Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.12 Parabola and Sine Path Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.13 Filling and Drawing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.14 Shading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.15 Specifying Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.16 Adding Arrow Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.17 Scoping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.18 Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.19 Repeating Things: For-Loops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.20 Adding Text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Editor of this documentation. Parts of this documentation have been written by other authors as indicated in these parts
or chapters and in Section 1.5.
3
3 Tutorial: A Petri-Net for Hagen 35
3.1 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2 Setting up the Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2.1 Setting up the Environment in L
A
T
E
X . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2.2 Setting up the Environment in Plain T
E
X . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2.3 Setting up the Environment in ConT
E
Xt . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.3 Introduction to Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.4 Placing Nodes Using the At Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.5 Using Styles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.6 Node Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.7 Naming Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.8 Placing Nodes Using Relative Placement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.9 Adding Labels Next to Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.10 Connecting Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.11 Adding Labels Next to Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.12 Adding the Snaked Line and Multi-Line Text . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.13 Using Layers: The Background Rectangles . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.14 The Complete Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4 Guidelines on Graphics 46
4.1 Planning the Time Needed for the Creation of Graphics . . . . . . . . . . . . . . . . . . . . 46
4.2 Workflow for Creating a Graphic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.3 Linking Graphics With the Main Text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.4 Consistency Between Graphics and Text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.5 Labels in Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.6 Plots and Charts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.7 Attention and Distraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
II Installation and Configuration 53
5 Installation 54
5.1 Package and Driver Versions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5.2 Installing Prebundled Packages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5.2.1 Debian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5.2.2 MiKTeX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
5.3 Installation in a texmf Tree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
5.3.1 Installation that Keeps Everything Together . . . . . . . . . . . . . . . . . . . . . . . 55
5.3.2 Installation that is TDS-Compliant . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
5.4 Updating the Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
6 Licenses and Copyright 56
6.1 Which License Applies? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
6.2 The GNU Public License, Version 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
6.2.1 Preamble . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
6.2.2 Terms and Conditions For Copying, Distribution and Modification . . . . . . . . . . 57
6.2.3 No Warranty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
6.3 The L
A
T
E
X Project Public License, Version 1.3c 2006-05-20 . . . . . . . . . . . . . . . . . . . 59
6.3.1 Preamble . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
6.3.2 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
6.3.3 Conditions on Distribution and Modification . . . . . . . . . . . . . . . . . . . . . . . 60
6.3.4 No Warranty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
6.3.5 Maintenance of The Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
6.3.6 Whether and How to Distribute Works under This License . . . . . . . . . . . . . . . 62
6.3.7 Choosing This License or Another License . . . . . . . . . . . . . . . . . . . . . . . . 62
6.3.8 A Recommendation on Modification Without Distribution . . . . . . . . . . . . . . . 63
6.3.9 How to Use This License . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
6.3.10 Derived Works That Are Not Replacements . . . . . . . . . . . . . . . . . . . . . . . 63
6.3.11 Important Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4
6.4 GNU Free Documentation License, Version 1.2, November 2002 . . . . . . . . . . . . . . . . 64
6.4.1 Preamble . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
6.4.2 Applicability and definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
6.4.3 Verbatim Copying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
6.4.4 Copying in Quantity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
6.4.5 Modifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
6.4.6 Combining Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
6.4.7 Collection of Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
6.4.8 Aggregating with independent Works . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
6.4.9 Translation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
6.4.10 Termination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
6.4.11 Future Revisions of this License . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
6.4.12 Addendum: How to use this License for your documents . . . . . . . . . . . . . . . . 68
7 Input and Output Formats 69
7.1 Supported Input Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
7.1.1 Using the L
A
T
E
X Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
7.1.2 Using the Plain T
E
X Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
7.1.3 Using the ConT
E
Xt Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
7.2 Supported Output Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
7.2.1 Selecting the Backend Driver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
7.2.2 Producing PDF Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
7.2.3 Producing PostScript Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
7.2.4 Producing HTML / SVG Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
7.2.5 Producing Perfectly Portable DVI Output . . . . . . . . . . . . . . . . . . . . . . . . 72
III TikZ ist kein Zeichenprogramm 73
8 Design Principles 74
8.1 Special Syntax For Specifying Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
8.2 Special Syntax For Path Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
8.3 Actions on Paths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
8.4 Key-Value Syntax for Graphic Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
8.5 Special Syntax for Specifying Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
8.6 Special Syntax for Specifying Trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
8.7 Grouping of Graphic Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
8.8 Coordinate Transformation System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
9 Hierarchical Structures: Package, Environments, Scopes, and Styles 77
9.1 Loading the Package and the Libraries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
9.2 Creating a Picture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
9.2.1 Creating a Picture Using an Environment . . . . . . . . . . . . . . . . . . . . . . . . 77
9.2.2 Creating a Picture Using a Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
9.2.3 Adding a Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
9.3 Using Scopes to Structure a Picture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
9.4 Using Scopes Inside Paths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
9.5 Using Styles to Manage How Pictures Look . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
10 Specifying Coordinates 82
10.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
10.2 Coordinate Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
10.2.1 Canvas, XYZ, and Polar Coordinate Systems . . . . . . . . . . . . . . . . . . . . . . 82
10.2.2 Barycentric Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
10.2.3 Node Coordinate System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
10.2.4 Intersection Coordinate Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
10.2.5 Defining New Coordinate Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
10.3 Relative and Incremental Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
5
11 Syntax for Path Specifications 90
11.1 The Move-To Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
11.2 The Line-To Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
11.2.1 Straight Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
11.2.2 Horizontal and Vertical Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
11.2.3 Snaked Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
11.3 The Curve-To Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
11.4 The Cycle Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
11.5 The Rectangle Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
11.6 Rounding Corners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
11.7 The Circle and Ellipse Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
11.8 The Arc Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
11.9 The Grid Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
11.10 The Parabola Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
11.11 The Sine and Cosine Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
11.12 The Plot Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
11.13 The To Path Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
11.14 The Scoping Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
11.15 The Node Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
11.16 The PGF-Extra Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
12 Actions on Paths 104
12.1 Specifying a Color . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
12.2 Drawing a Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
12.2.1 Graphic Parameters: Line Width, Line Cap, and Line Join . . . . . . . . . . . . . . . 106
12.2.2 Graphic Parameters: Dash Pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
12.2.3 Graphic Parameters: Draw Opacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
12.2.4 Graphic Parameters: Arrow Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
12.2.5 Graphic Parameters: Double Lines and Bordered Lines . . . . . . . . . . . . . . . . . 110
12.3 Filling a Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
12.3.1 Graphic Parameters: Fill Pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
12.3.2 Graphic Parameters: Interior Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
12.3.3 Graphic Parameters: Fill Opacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
12.4 Shading a Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
12.4.1 Choosing a Shading Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
12.4.2 Choosing a Shading Color . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
12.5 Establishing a Bounding Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
12.6 Using a Path For Clipping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
13 Nodes and Edges 118
13.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
13.2 Nodes and Their Shapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
13.3 Multi-Part Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
13.4 Options for the Text in Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
13.5 Placing Nodes Using Anchors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
13.6 Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
13.7 Placing Nodes on a Line or Curve Explicitly . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
13.8 Placing Nodes on a Line or Curve Implicitly . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
13.9 The Label and Pin Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
13.10 Connecting Nodes: Using Nodes as Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . 131
13.11 Connecting Nodes: Using the Edge Operation . . . . . . . . . . . . . . . . . . . . . . . . . . 131
13.12 Referencing Nodes Outside the Current Pictures . . . . . . . . . . . . . . . . . . . . . . . . . 133
13.12.1Referencing a Node in a Different Picture . . . . . . . . . . . . . . . . . . . . . . . . 133
13.12.2Referencing the Current Page Node – Absolute Positioning . . . . . . . . . . . . . . . 134
13.13 Predefined Shapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
13.14 Executing Code After Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
6
14 Matrices and Alignment 138
14.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
14.2 Matrices are Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
14.3 Cell Pictures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
14.3.1 Alignment of Cell Pictures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
14.3.2 Setting and Adjusting Column and Row Spacing . . . . . . . . . . . . . . . . . . . . 140
14.3.3 Cell Styles and Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
14.4 Anchoring a Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
14.5 Considerations Concerning Active Characters . . . . . . . . . . . . . . . . . . . . . . . . . . 144
14.6 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
15 Making Trees Grow 148
15.1 Introduction to the Child Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
15.2 Child Paths and the Child Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
15.3 Naming Child Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
15.4 Specifying Options for Trees and Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
15.5 Placing Child Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
15.6 Edges From the Parent Node . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
16 Plots of Functions 158
16.1 When Should One Use TikZ for Generating Plots? . . . . . . . . . . . . . . . . . . . . . . . 158
16.2 The Plot Path Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
16.3 Plotting Points Given Inline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
16.4 Plotting Points Read From an External File . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
16.5 Plotting a Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
16.6 Plotting a Function Using Gnuplot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
16.7 Placing Marks on the Plot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
16.8 Smooth Plots, Sharp Plots, and Comb Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
17 Transformations 167
17.1 The Different Coordinate Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
17.2 The XY- and XYZ-Coordinate Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
17.3 Coordinate Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
IV Libraries 172
18 Arrow Tip Library 173
18.1 Triangular Arrow Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
18.2 Barbed Arrow Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
18.3 Bracket-Like Arrow Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
18.4 Circle and Diamond Arrow Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
18.5 Serif-Like Arrow Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
18.6 Partial Arrow Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
18.7 Line Caps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
19 Automata Drawing Library 175
19.1 Drawing Automata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
19.2 States With and Without Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
19.3 Initial and Accepting States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
19.4 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
20 Background Library 180
7
21 Calendar Library 183
21.1 Calendar Command . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
21.1.1 Creating a Simple List of Days . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
21.1.2 Adding a Month Label . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
21.1.3 Creating a Week List Arrangement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
21.1.4 Creating a Month List Arrangement . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
21.2 Arrangements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
21.3 Month Labels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
21.4 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
22 Entity-Relationship Diagram Drawing Library 199
22.1 Entities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
22.2 Relationships . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
22.3 Attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
23 Paper Folding Diagrams Library 201
24 Matrix Library 204
24.1 Matrices of Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
24.2 Delimiters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
25 Mindmap Drawing Library 207
25.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
25.2 The Mindmap Style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
25.3 Concepts Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
25.3.1 Isolated Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
25.3.2 Concepts in Trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
25.4 Connecting Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
25.4.1 Simple Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
25.4.2 The Circle Connection Bar Snake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
25.4.3 The Circle Connection Bar To-Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
25.4.4 Tree Edges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
25.5 Adding Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
25.6 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
26 Pattern Library 217
27 Petri-Net Drawing Library 218
27.1 Places . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
27.2 Transitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
27.3 Tokens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
28 Plot Handler Library 221
28.1 Curve Plot Handlers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
28.2 Comb Plot Handlers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
28.3 Mark Plot Handler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
29 Plot Mark Library 225
30 Shape Library 226
30.1 Geometric Shapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
30.2 Symbol Shapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
30.3 Shapes with Multiple Text Parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
30.4 Miscellaneous Shapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
31 Snake Library 235
8
32 To Path Library 238
32.1 Straight Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
32.2 Curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
32.3 Loops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
33 Tree Library 242
33.1 Growth Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
33.2 Edges From Parent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
V Utilities 245
34 Repeating Things: The Foreach Statement 246
35 Date and Calendar Utility Macros 250
35.1 Handling Dates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
35.1.1 Conversions Between Date Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
35.1.2 Checking Dates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
35.1.3 Typesetting Dates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
35.1.4 Localization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
35.2 Typesetting Calendars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
36 Page Management 256
36.1 Basic Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
36.2 The Predefined Layouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
36.3 Defining a Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
36.4 Creating Logical Pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262
37 Extended Color Support 263
VI Mathematical Engine 264
38 Design Principles 265
38.1 Loading the Mathematical Engine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
38.2 Layers of the Mathematical Engine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
38.3 Efficiency and Accuracy of the Mathematical Engine . . . . . . . . . . . . . . . . . . . . . . 265
39 Evaluating Mathematical Expressions 266
39.1 Commands for Parsing Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
39.2 Syntax for mathematical expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
40 Evaluating Mathematical Operations 271
40.1 Basic Operations and Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
40.2 Trignometric Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
40.3 Pseudo-Random Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
40.4 Conversion Between Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
41 Reimplementing the Computations of the Mathematical Engine 275
VII The Basic Layer 276
42 Design Principles 277
42.1 Core and Optional Packages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
42.2 Communicating with the Basic Layer via Macros . . . . . . . . . . . . . . . . . . . . . . . . 278
42.3 Path-Centered Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
42.4 Coordinate Versus Canvas Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
9
43 Hierarchical Structures: Package, Environments, Scopes, and Text 280
43.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
43.1.1 The Hierarchical Structure of the Package . . . . . . . . . . . . . . . . . . . . . . . . 280
43.1.2 The Hierarchical Structure of Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . 280
43.2 The Hierarchical Structure of the Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
43.2.1 The Main Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
43.2.2 The Core Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
43.2.3 The Optional Basic Layer Packages . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
43.2.4 The Library Packages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
43.3 The Hierarchical Structure of the Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
43.3.1 The Main Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
43.3.2 Graphic Scope Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
43.3.3 Inserting Text and Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
44 Specifying Coordinates 289
44.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
44.2 Basic Coordinate Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
44.3 Coordinates in the XY-Coordinate System . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
44.4 Three Dimensional Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
44.5 Building Coordinates From Other Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . 291
44.5.1 Basic Manipulations of Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
44.5.2 Points Traveling along Lines and Curves . . . . . . . . . . . . . . . . . . . . . . . . . 292
44.5.3 Points on Borders of Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
44.5.4 Points on the Intersection of Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
44.6 Extracting Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
44.7 Internals of How Point Commands Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
45 Constructing Paths 296
45.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
45.2 The Move-To Path Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
45.3 The Line-To Path Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
45.4 The Curve-To Path Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
45.5 The Close Path Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
45.6 Arc, Ellipse and Circle Path Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
45.7 Rectangle Path Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
45.8 The Grid Path Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
45.9 The Parabola Path Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
45.10 Sine and Cosine Path Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
45.11 Plot Path Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
45.12 Rounded Corners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
45.13 Internal Tracking of Bounding Boxes for Paths and Pictures . . . . . . . . . . . . . . . . . . 303
46 Snakes 305
46.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
46.2 Declaring a Snake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
46.2.1 Segments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
46.2.2 Snake Automata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
46.2.3 The Snake Declaration Command . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
46.2.4 Predefined Snakes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
46.3 Using Snakes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
47 Using Paths 310
47.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
47.2 Stroking a Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
47.2.1 Graphic Parameter: Line Width . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
47.2.2 Graphic Parameter: Caps and Joins . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
47.2.3 Graphic Parameter: Dashing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
47.2.4 Graphic Parameter: Stroke Color . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
47.2.5 Graphic Parameter: Stroke Opacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
10
47.2.6 Graphic Parameter: Arrows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
47.3 Filling a Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
47.3.1 Graphic Parameter: Interior Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
47.3.2 Graphic Parameter: Filling Color . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
47.3.3 Graphic Parameter: Fill Opacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
47.4 Clipping a Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
47.5 Using a Path as a Bounding Box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
48 Arrow Tips 316
48.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
48.1.1 When Does PGF Draw Arrow Tips? . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
48.1.2 Meta-Arrow Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
48.2 Declaring an Arrow Tip Kind . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
48.3 Declaring a Derived Arrow Tip Kind . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
48.4 Using an Arrow Tip Kind . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
48.5 Predefined Arrow Tip Kinds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
49 Nodes and Shapes 323
49.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
49.1.1 Creating and Referencing Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
49.1.2 Anchors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
49.1.3 Layers of a Shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
49.1.4 Node Parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324
49.2 Creating Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324
49.3 Using Anchors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
49.3.1 Referencing Anchors of Nodes in the Same Picture . . . . . . . . . . . . . . . . . . . 327
49.3.2 Referencing Anchors of Nodes in Different Pictures . . . . . . . . . . . . . . . . . . . 328
49.4 Predefined Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328
49.5 Declaring New Shapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
49.5.1 What Must Be Defined For a Shape? . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
49.5.2 Normal Anchors Versus Saved Anchors . . . . . . . . . . . . . . . . . . . . . . . . . . 329
49.5.3 Command for Declaring New Shapes . . . . . . . . . . . . . . . . . . . . . . . . . . . 330
49.6 Predefined Shapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
50 Matrices 337
50.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
50.2 Cell Pictures and Their Alignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
50.3 The Matrix Command . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
50.4 Row and Column Spacing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
50.5 Callbacks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
51 Coordinate and Canvas Transformations 342
51.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
51.2 Coordinate Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
51.2.1 How PGF Keeps Track of the Coordinate Transformation Matrix . . . . . . . . . . . 342
51.2.2 Commands for Relative Coordinate Transformations . . . . . . . . . . . . . . . . . . 342
51.2.3 Commands for Absolute Coordinate Transformations . . . . . . . . . . . . . . . . . . 346
51.2.4 Saving and Restoring the Coordinate Transformation Matrix . . . . . . . . . . . . . . 347
51.3 Canvas Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
52 Patterns 349
52.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
52.2 Declaring a Pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
52.3 Setting a Pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350
11
53 Declaring and Using Images 351
53.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
53.2 Declaring an Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
53.3 Using an Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
53.4 Masking an Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
54 Externalizing Graphics 355
54.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
54.2 Workflow Step 1: Naming Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
54.3 Workflow Step 2: Generating the External Graphics . . . . . . . . . . . . . . . . . . . . . . . 356
54.4 Workflow Step 3: Including the External Graphics . . . . . . . . . . . . . . . . . . . . . . . . 357
54.5 A Complete Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358
55 Declaring and Using Shadings 360
55.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360
55.2 Declaring Shadings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360
55.3 Using Shadings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361
56 Creating Plots 365
56.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
56.2 Generating Plot Streams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
56.2.1 Basic Building Blocks of Plot Streams . . . . . . . . . . . . . . . . . . . . . . . . . . 365
56.2.2 Commands That Generate Plot Streams . . . . . . . . . . . . . . . . . . . . . . . . . 366
56.3 Plot Handlers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368
57 Layered Graphics 369
57.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
57.2 Declaring Layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
57.3 Using Layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
58 Quick Commands 371
58.1 Quick Coordiante Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371
58.2 Quick Path Construction Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371
58.3 Quick Path Usage Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
58.4 Quick Text Box Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
VIII The System Layer 373
59 Design of the System Layer 374
59.1 Driver Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
59.2 Common Definition Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
60 Commands of the System Layer 375
60.1 Beginning and Ending a Stream of System Commands . . . . . . . . . . . . . . . . . . . . . 375
60.2 Path Construction System Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376
60.3 Canvas Transformation System Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
60.4 Stroking, Filling, and Clipping System Commands . . . . . . . . . . . . . . . . . . . . . . . . 377
60.5 Graphic State Option System Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
60.6 Color System Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
60.7 Pattern System Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
60.8 Scoping System Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
60.9 Image System Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382
60.10 Shading System Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383
60.11 Reusable Objects System Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383
60.12 Invisibility System Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383
60.13 Position Tracking Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
60.14 Internal Conversion Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
12
61 The Soft Path Subsystem 386
61.1 Path Creation Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386
61.2 Starting and Ending a Soft Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386
61.3 Soft Path Creation Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
61.4 The Soft Path Data Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
62 The Protocol Subsystem 389
IX References and Index 390
Index 391
13
1 Introduction
The pgf package, where “pgf” is supposed to mean “portable graphics format” (or “pretty, good, functional”
if you prefer. . . ), is a package for creating graphics in an “inline” manner. It defines a number of T
E
X
commands that draw graphics. For example, the code \tikz \draw (0pt,0pt) -- (20pt,6pt); yields the
line and the code \tikz \fill[orange] (1ex,1ex) circle (1ex); yields .
In a sense, when you use pgf you “program” your graphics, just as you “program” your document when
you use T
E
X. You get all the advantages of the “T
E
X-approach to typesetting” for your graphics: quick
creation of simple graphics, precise positioning, the use of macros, often superior typography. You also
inherit all the disadvantages: steep learning curve, no wysiwyg, small changes require a long recompilation
time, and the code does not really “show” how things will look like.
1.1 Structure of the System
The pgf system consists of different layers:
System layer: This layer provides a complete abstraction of what is going on “in the driver.” The driver
is a program like dvips or dvipdfm that takes a .dvi file as input and generates a .ps or a .pdf file.
(The pdftex program also counts as a driver, even though it does not take a .dvi file as input. Never
mind.) Each driver has its own syntax for the generation of graphics, causing headaches to everyone
who wants to create graphics in a portable way. pgf’s system layer “abstracts away” these differences.
For example, the system command \pgfsys@lineto{10pt}{10pt} extends the current path to the
coordinate (10pt, 10pt) of the current {pgfpicture}. Depending on whether dvips, dvipdfm, or
pdftex is used to process the document, the system command will be converted to different \special
commands. The system layer is as “minimalistic” as possible since each additional command makes it
more work to port pgf to a new driver.
As a user, you will not use the system layer directly.
Basic layer: The basic layer provides a set of basic commands that allow you to produce complex graphics
in a much easier manner than by using the system layer directly. For example, the system layer provides
no commands for creating circles since circles can be composed from the more basic B´ezier curves (well,
almost). However, as a user you will want to have a simple command to create circles (at least I do)
instead of having to write down half a page of B´ezier curve support coordinates. Thus, the basic layer
provides a command \pgfpathcircle that generates the necessary curve coordinates for you.
The basic layer is consists of a core, which consists of several interdependent packages that can only
be loaded en bloc, and additional packages that extend the core by more special-purpose commands
like node management or a plotting interface. For instance, the beamer package uses the core, but
not all of the additional packages of the basic layer.
Frontend layer: A frontend (of which there can be several) is a set of commands or a special syntax that
makes using the basic layer easier. A problem with directly using the basic layer is that code written
for this layer is often too “verbose.” For example, to draw a simple triangle, you may need as many as
five commands when using the basic layer: One for beginning a path at the first corner of the triangle,
one for extending the path to the second corner, one for going to the third, one for closing the path,
and one for actually painting the triangle (as opposed to filling it). With the tikz frontend all this
boils down to a single simple metafont-like command:
\draw (0,0) -- (1,0) -- (1,1) -- cycle;
There are different frontends:
• The TikZ frontend is the “natural” frontend for pgf. It gives you access to all features of pgf,
but it is intended to be easy to use. The syntax is a mixture of metafont and pstricks and
some ideas of myself. This frontend is neither a complete metafont compatibility layer nor a
pstricks compatibility layer and it is not intended to become either.
• The pgfpict2e frontend reimplements the standard L
A
T
E
X {picture} environment and com-
mands like \line or \vector using the pgf basic layer. This layer is not really “necessary” since
the pict2e.sty package does at least as good a job at reimplementing the {picture} environ-
ment. Rather, the idea behind this package is to have a simple demonstration of how a frontend
can be implemented.
14
It would be possible to implement a pgftricks frontend that maps pstricks commands to pgf
commands. However, I have not done this and even if fully implemented, many things that work in
pstricks will not work, namely whenever some pstricks command relies too heavily on PostScript
trickery. Nevertheless, such a package might be useful in some situations.
As a user of pgf you will use the commands of a frontend plus perhaps some commands of the basic
layer. For this reason, this manual explains the frontends first, then the basic layer, and finally the system
layer.
1.2 Comparison with Other Graphics Packages
pgf is not the only graphics package for T
E
X. In the following, I try to give a reasonably fair comparison of
the pgf-system and other packages.
1. The standard L
A
T
E
X {picture} environment allows you to create simple graphics, but little more. This
is certainly not due to a lack of knowledge or imagination on the part of L
A
T
E
X’s designer(s). Rather,
this is the price paid for the {picture} environment’s portability: It works together with all backend
drivers.
2. The pstricks package is certainly powerful enough to create any conceivable kind of graphic, but it
is not portable at all. Most importantly, it does not work with pdftex nor with any other driver that
produces anything but PostScript code.
Compared to pgf, pstricks has a broader support base. There are many nice extra packages for
special purpose sitations that have been contributed by users over the last decade.
The TikZ syntax is more consistent than the pstricks syntax as TikZ was developed “in a more
centralized manner” and also “with the shortcomings on pstricks in mind.”
Note that a number of neat tricks that pstricks can do are impossible in pgf. In particular, pstricks
has access to the powerful PostScript programming language, which allows trickery such as inline
function plotting.
3. The xypic package is an older package for creating graphics. However, it is more difficult to use and
to learn because the syntax and the documentation are a bit cryptic.
4. The dratex package is a small graphic package for creating a graphics. Compared to the other package,
including pgf, it is very small, which may or may not be an advantage.
5. The metapost program is a very powerful alternative to pgf. However, it is an external program,
which entails a bunch of problems. The time needed both to create a small graphic and also to compile
it is much greater than in pgf. The main problem with metapost, however, is the inclusion of labels.
This is much easier to achieve using pgf.
6. The xfig program is an important alternative to TikZ for users who do not wish to “program” their
graphics as is necessary with TikZ and the other packages above. Their is a conversion program that
will convert xfig graphics to both TikZ and for pgf, but it is still under construction.
1.3 Utilities: Page Management
The pgf package include a special subpackage called pgfpages, which is used to assemble several pages into
a single page. This package is not really about creating graphics, but it is part of pgf nevertheless, mostly
because its implementation uses pgf heavily.
The subpackage pgfpages provides commands for assembling several “virtual pages” into a single “phys-
ical page.” The idea is that whenever T
E
X has a page ready for “shipout,” pgfpages interrupts this shipout
and instead stores the page to be shipped out in a special box. When enough “virtual pages” have been
accumulated in this way, they are scaled down and arranged on a “physical page,” which then really shipped
out. This mechanism allows you to create “two page on one page” versions of a document directly inside
L
A
T
E
X without the use of any external programs.
However, pgfpages can do quite a lot more than that. You can use it to put logos and watermark on
pages, print up to 16 pages on one page, add borders to pages, and more.
15
1.4 How to Read This Manual
This manual describes both the design of the pgf system and its usage. The organization is very roughly
according to “user-friendliness.” The commands and subpackages that are easiest and most frequently used
are described first, more low-level and esoteric features are discussed later.
If you have not yet installed pgf, please read the installation first. Second, it might be a good idea to
read the tutorial. Finally, you might wish to skim through the description of TikZ. Typically, you will not
need to read the sections on the basic layer. You will only need to read the part on the system layer if you
intend to write your own frontend or if you wish to port pgf to a new driver.
The “public” commands and environments provided by the pgf package are described throughout the
text. In each such description, the described command, environment or option is printed in red. Text shown
in green is optional and can be left out.
1.5 Authors and Acknowledgements
The bulk of the pgf system and its documentation was written by Till Tantau. The pgf mathematical
engine was written and documented by Mark Wibrow. Additionally, numerous people have contributed to
the pgf system by writing emails, spotting bugs, or sending libraries. Many thanks to all these people, who
are too numerous to name them all!
1.6 Getting Help
When you need help with pgf and TikZ, please do the following:
1. Read the manual, at least the part that has to do with your problem.
2. If that does not solve the problem, try having a look at the sourceforge development page for pgf and
TikZ (see the title of this document). Perhaps someone has already reported a similar problem and
someone has found a solution.
3. On the website you will find numerous forums for getting help. There, you can write to help forums,
file bug reports, join mailing lists, and so on.
4. Before you file a bug report, especially a bug report concerning the installation, make sure that this
is really a bug. In particular, have a look at the .log file that results when you T
E
X your files. This
.log file should show that all the right files are loaded from the right directories. Nearly all installation
problems can be resolved by looking at the .log file.
5. As a last resort you can try to email me (Till Tantau) or, if the problem concerns the mathematical
engine, Mark Wibrow. I do not mind getting emails, I simply get way too many of them. Because of
this, I cannot guarantee that your emails will be answered timely or even at all. Your chances that
your problem will be fixed are somewhat higher if you mail to the pgf mailing list (naturally, I read
this list and answer questions when I have the time).
6. Please, do not phone me in my office (unless, of course, you attend one of my lectures).
16
Part I
Tutorials and Guidelines
by Till Tantau
To help you get started with TikZ, instead of a long installation and configuration section, this manual starts
with tutorials. They explain all the basic and some of the more advanced features of the system, without
going into all the details. This part also contains some guidelines on how you should proceed when creating
graphics using TikZ.
\tikz \draw[thick,rounded corners=8pt]
(0,0) -- (0,2) -- (1,3.25) -- (2,2) -- (2,0) -- (0,2) -- (2,2) -- (0,0) -- (2,0);
17
2 Tutorial: A Picture for Karl’s Students
This tutorial is intended for new users of pgf and TikZ. It does not give an exhaustive account of all the
features of TikZ or pgf, just of those that you are likely to use right away.
Karl is a math and chemistry high-school teacher. He used to create the graphics in his worksheets and
exams using L
A
T
E
X’s {picture} environment. While the results were acceptable, creating the graphics often
turned out to be a lengthy process. Also, there tended to be problems with lines having slightly wrong angles
and circles also seemed to be hard to get right. Naturally, his students could not care less whether the lines
had the exact right angles and they find Karl’s exams too difficult no matter how nicely they were drawn.
But Karl was never entirely satisfied with the result.
Karl’s son, who was even less satisfied with the results (he did not have to take the exams, after all),
told Karl that he might wish to try out a new package for creating graphics. A bit confusingly, this package
seems to have two names: First, Karl had to download and install a package called pgf. Then it turns out
that inside this package there is another package called TikZ, which is supposed to stand for “TikZ ist kein
Zeichenprogramm.” Karl finds this all a bit strange and TikZ seems to indicate that the package does not
do what he needs. However, having used gnu software for quite some time and “gnu not being Unix,” there
seems to be hope yet. His son assures him that TikZ’s name is intended to warn people that TikZ is not a
program that you can use to draw graphics with your mouse or tablet. Rather, it is more like a “graphics
language.”
2.1 Problem Statement
Karl wants to put a graphic on the next worksheet for his students. He is currently teaching his students
about sine and cosine. What he would like to have is something that looks like this (ideally):
x
y
−1

1
2
1
−1

1
2
1
2
1
α
sin α
cos α
tan α =
sin α
cos α
The angle α is 30

in the example
(π/6 in radians). The sine of α, which
is the height of the red line, is
sin α = 1/2.
By the Theorem of Pythagoras we
have cos
2
α + sin
2
α = 1. Thus the
length of the blue line, which is the
cosine of α, must be
cos α =

1 − 1/4 =
1
2

3.
This shows that tan α, which is the
height of the orange line, is
tan α =
sin α
cos α
= 1/

3.
2.2 Setting up the Environment
In TikZ, to draw a picture, at the start of the picture you need to tell T
E
X or L
A
T
E
X that you want to
start a picture. In L
A
T
E
X this is done using the environment {tikzpicture}, in plain T
E
X you just use
\tikzpicture to start the picture and \endtikzpicture to end it.
2.2.1 Setting up the Environment in L
A
T
E
X
Karl, being a L
A
T
E
X user, thus sets up his file as follows:
18
\documentclass{article} % say
\usepackage{tikz}
\begin{document}
We are working on
\begin{tikzpicture}
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\end{tikzpicture}.
\end{document}
When executed, that is, run via pdflatex or via latex followed by dvips, the resulting will contain
something that looks like this:
We are working on .
We are working on
\begin{tikzpicture}
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\end{tikzpicture}.
Admittedly, not quite the whole picture, yet, but we do have the axes established. Well, not quite, but
we have the lines that make up the axes drawn. Karl suddenly has a sinking feeling that the picture is still
some way off.
Let’s have a more detailed look at the code. First, the package tikz is loaded. This package is a so-called
“frontend” to the basic pgf system. The basic layer, which is also described in this manual, is somewhat
more, well, basic and thus harder to use. The frontend makes things easier by providing a simpler syntax.
Inside the environment there are two \draw commands. They mean: “The path, which is specified follow-
ing the command up to the semicolon, should be drawn.” The first path is specified as (-1.5,0) -- (0,1.5),
which means “a straight line from the point at position (−1.5, 0) to the point at position (0, 1.5).” Here, the
positions are specified within a special coordinate system in which, initially, one unit is 1cm.
Karl is quite pleased to note that the environment automatically reserves enough space to encompass the
picture.
2.2.2 Setting up the Environment in Plain T
E
X
Karl’s wife Gerda, who also happens to be a math teacher, is not a L
A
T
E
X user, but uses plain T
E
X since
she prefers to do things “the old way.” She can also use TikZ. Instead of \usepackage{tikz} she has
to write \input tikz.tex and instead of \begin{tikzpicture} she writes \tikzpicture and instead of
\end{tikzpicture} she writes \endtikzpicture.
Thus, she would use:
%% Plain TeX file
\input tikz.tex
\baselineskip=12pt
\hsize=6.3truein
\vsize=8.7truein
We are working on
\tikzpicture
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\endtikzpicture.
\bye
Gerda can typeset this file using either pdftex or tex together with dvips. TikZ will automatically
discern which driver she is using. If she wishes to use dvipdfm together with tex, she either needs to
modify the file pgf.cfg or can write \def\pgfsysdriver{pgfsys-dvipdfm.def} somewhere before she
inputs tikz.tex or pgf.tex.
2.2.3 Setting up the Environment in ConT
E
Xt
Karl’s uncle Hans uses ConT
E
Xt. Like Gerda, Hans can also use TikZ. Instead of \usepackage{tikz} he
says \usemodule[tikz]. Instead of \begin{tikzpicture} he writes \starttikzpicture and instead of
\end{tikzpicture} he writes \stoptikzpicture.
19
His version of the example looks like this:
%% ConTeXt file
\usemodule[tikz]
\starttext
We are working on
\starttikzpicture
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\stoptikzpicture.
\stoptext
Hans will now typeset this file in the usual way using texexec.
2.3 Straight Path Construction
The basic building block of all pictures in TikZ is the path. A path is a series of straight lines and curves
that are connected (that is not the whole picture, but let us ignore the complications for the moment). You
start a path by specifying the coordinates of the start position as a point in round brackets, as in (0,0).
This is followed by a series of “path extension operations.” The simplest is --, which we used already. It
must be followed by another coordinate and it extends the path in a straight line to this new position. For
example, if we were to turn the two paths of the axes into one path, the following would result:
\tikz \draw (-1.5,0) -- (1.5,0) -- (0,-1.5) -- (0,1.5);
Karl is a bit confused by the fact that there is no {tikzpicture} environment, here. Instead, the little
command \tikz is used. This command either takes one argument (starting with an opening brace as in
\tikz{\draw (0,0) -- (1.5,0)}, which yields ) or collects everything up to the next semicolon
and puts it inside a {tikzpicture} environment. As a rule of thumb, all TikZ graphic drawing commands
must occur as an argument of \tikz or inside a {tikzpicture} environment. Fortunately, the command
\draw will only be defined inside this environment, so there is little chance that you will accidentally do
something wrong here.
2.4 Curved Path Construction
The next thing Karl wants to do is to draw the circle. For this, straight lines obviously will not do. Instead,
we need some way to draw curves. For this, TikZ provides a special syntax. One or two “control points”
are needed. The math behind them is not quite trivial, but here is the basic idea: Suppose you are at point
x and the first control point is y. Then the curve will start “going in the direction of y at x,” that is, the
tangent of the curve at x will point toward y. Next, suppose the curve should end at z and the second
support point is w. Then the curve will, indeed, end at z and the tangent of the curve at point z will go
through w.
Here is an example (the control points have been added for clarity):
\begin{tikzpicture}
\filldraw [gray] (0,0) circle (2pt)
(1,1) circle (2pt)
(2,1) circle (2pt)
(2,0) circle (2pt);
\draw (0,0) .. controls (1,1) and (2,1) .. (2,0);
\end{tikzpicture}
The general syntax for extending a path in a “curved” way is .. controls ¸first control point) and
¸second control point) .. ¸end point). You can leave out the and ¸second control point), which causes the
first one to be used twice.
So, Karl can now add the first half circle to the picture:
20
\begin{tikzpicture}
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (-1,0) .. controls (-1,0.555) and (-0.555,1) .. (0,1)
.. controls (0.555,1) and (1,0.555) .. (1,0);
\end{tikzpicture}
Karl is happy with the result, but finds specifying circles in this way to be extremely awkward. Fortu-
nately, there is a much simpler way.
2.5 Circle Path Construction
In order to draw a circle, the path construction operation circle can be used. This operation is followed
by a radius in round brackets as in the following example: (Note that the previous position is used as the
center of the circle.)
\tikz \draw (0,0) circle (10pt);
You can also append an ellipse to the path using the ellipse operation. Instead of a single radius you
can specify two of them, one for the x-direction and one for the y-direction, separated by and:
\tikz \draw (0,0) ellipse (20pt and 10pt);
To draw an ellipse whose axes are not horizontal and vertical, but point in an arbitrary direction (a
“turned ellipse” like ) you can use transformations, which are explained later. The code for the little
ellipse is \tikz \draw[rotate=30] (0,0) ellipse (6pt and 3pt);, by the way.
So, returning to Karl’s problem, he can write \draw (0,0) circle (1cm); to draw the circle:
\begin{tikzpicture}
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\end{tikzpicture}
At this point, Karl is a bit alarmed that the circle is so small when he wants the final picture to be much
bigger. He is pleased to learn that TikZ has powerful transformation options and scaling everything by a
factor of three is very easy. But let us leave the size as it is for the moment to save some space.
2.6 Rectangle Path Construction
The next things we would like to have is the grid in the background. There are several ways to produce it.
For example, one might draw lots of rectangles. Since rectangles are so common, there is a special syntax
for them: To add a rectangle to the current path, use the rectangle path construction operation. This
operation should be followed by another coordinate and will append a rectangle to the path such that the
previous coordinate and the next coordinates are corners of the rectangle. So, let us add two rectangles to
the picture:
\begin{tikzpicture}
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\draw (0,0) rectangle (0.5,0.5);
\draw (-0.5,-0.5) rectangle (-1,-1);
\end{tikzpicture}
21
While this may be nice in other situations, this is not really leading anywhere with Karl’s problem: First,
we would need an awful lot of these rectangles and then there is the border that is not “closed.”
So, Karl is about to resort to simply drawing four vertical and four horizontal lines using the nice \draw
command, when he learns that there is a grid path construction operation.
2.7 Grid Path Construction
The grid path operation adds a grid to the current path. It will add lines making up a grid that fills
the rectangle whose one corner is the current point and whose other corner is the point following the grid
operation. For example, the code \tikz \draw[step=2pt] (0,0) grid (10pt,10pt); produces . Note
how the optional argument for \draw can be used to specify a grid width (there are also xstep and ystep to
define the steppings independently). As Karl will learn soon, there are lots of things that can be influenced
using such options.
For Karl, the following code could be used:
\begin{tikzpicture}
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\draw[step=.5cm] (-1.4,-1.4) grid (1.4,1.4);
\end{tikzpicture}
Having another look at the desired picture, Karl notices that it would be nice for the grid to be more
subdued. (His son told him that grids tend to be distracting if they are not subdued.) To subdue the grid,
Karl adds two more options to the \draw command that draws the grid. First, he uses the color gray for the
grid lines. Second, he reduces the line width to very thin. Finally, he swaps the ordering of the commands
so that the grid is drawn first and everything else on top.
\begin{tikzpicture}
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\end{tikzpicture}
2.8 Adding a Touch of Style
Instead of the options gray,very thin Karl could also have said style=help lines. Styles are predefined
sets of options that can be used to organize how a graphic is drawn. By saying style=help lines you say
“use the style that I (or someone else) has set for drawing help lines.” If Karl decides, at some later point,
that grids should be drawn, say, using the color blue!50 instead of gray, he could say the following:
\tikzstyle help lines=[color=blue!50,very thin]
Alternatively, he could have said the following:
\tikzstyle help lines+=[color=blue!50]
This would have added the color=blue!50 option. The help lines style would now contain two color
options, but the second would override the first.
Using styles makes your graphics code more flexible. You can change the way things look easily in a
consistent manner.
To build a hierarchy of styles you can have one style use another. So in order to define a style Karl’s grid
that is based on the grid style Karl could say
\tikzstyle Karl’s grid=[style=help lines,color=blue!50]
...
\draw[style=Karl’s grid] (0,0) grid (5,5);
22
You can also leave out the style=. Thus, whenever TikZ encounters an options that it does not know
about, it will check whether this option happens to be the name of a style. If so, the style is used. Thus,
Karl could also have written:
\tikzstyle Karl’s grid=[help lines,color=blue!50]
...
\draw[Karl’s grid] (0,0) grid (5,5);
For some styles, like the very thin style, it is pretty clear what the style does and there is no need to say
style=very thin. For other styles, like help lines, it seems more natural to me to say style=help lines.
But, mainly, this is a matter of taste.
2.9 Drawing Options
Karl wonders what other options there are that influence how a path is drawn. He saw already that the
color=¸color) option can be used to set the line’s color. The option draw=¸color) does nearly the same, only
it sets the color for the lines only and a different color can be used for filling (Karl will need this when he
fills the arc for the angle).
He saw that the style very thin yields very thin lines. Karl is not really surprised by this and neither
is he surprised to learn that thin yields thin lines, thick yields thick lines, very thick yields very thick
lines, ultra thick yields really, really thick lines and ultra thin yields lines that are so thin that low-
resolution printers and displays will have trouble showing them. He wonders what gives lines of “normal”
thickness. It turns out that thin is the correct choice. This seems strange to Karl, but his son explains
him that L
A
T
E
X has two commands called \thinlines and \thicklines and that \thinlines gives the line
width of “normal” lines, more precisely, of the thickness that, say, the stem of a letter like “T” or “i” has.
Nevertheless, Karl would like to know whether there is anything “in the middle” between thin and thick.
There is: semithick.
Another useful thing one can do with lines is to dash or dot them. For this, the two styles dashed and
dotted can be used, yielding and . Both options also exist in a loose and a dense version, called
loosely dashed, densely dashed, loosely dotted, and densely dotted. If he really, really needs to,
Karl can also define much more complex dashing patterns with the dash pattern option, but his son insists
that dashing is to be used with utmost care and mostly distracts. Karl’s son claims that complicated dashing
patterns are evil. Karl’s students do not care about dashing patterns.
2.10 Arc Path Construction
Our next obstacle is to draw the arc for the angle. For this, the arc path construction operation is useful,
which draws part of a circle or ellipse. This arc operation must be followed by a triple in rounded brackets,
where the components of the triple are separated by colons. The first two components are angles, the last
one is a radius. An example would be (10:80:10pt), which means “an arc from 10 degrees to 80 degrees
on a circle of radius 10pt.” Karl obviously needs an arc from 0

to 30

. The radius should be something
relatively small, perhaps around one third of the circle’s radius. This gives: (0:30:3mm).
When one uses the arc path construction operation, the specified arc will be added with its starting point
at the current position. So, we first have to “get there.”
\begin{tikzpicture}
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\draw (3mm,0mm) arc (0:30:3mm);
\end{tikzpicture}
Karl thinks this is really a bit small and he cannot continue unless he learns how to do scaling. For this,
he can add the [scale=3] option. He could add this option to each \draw command, but that would be
awkward. Instead, he adds it to the whole environment, which causes this option to apply to everything
within.
23
\begin{tikzpicture}[scale=3]
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\draw (3mm,0mm) arc (0:30:3mm);
\end{tikzpicture}
As for circles, you can specify “two” radii in order to get an elliptical arc.
\tikz \draw (0,0) arc (0:315:1.75cm and 1cm);
2.11 Clipping a Path
In order to save space in this manual, it would be nice to clip Karl’s graphics a bit so that we can focus on
the “interesting” parts. Clipping is pretty easy in TikZ. You can use the \clip command clip all subsequent
drawing. It works like \draw, only it does not draw anything, but uses the given path to clip everything
subsequently.
\begin{tikzpicture}[scale=3]
\clip (-0.1,-0.2) rectangle (1.1,0.75);
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\draw (3mm,0mm) arc (0:30:3mm);
\end{tikzpicture}
You can also do both at the same time: Draw and clip a path. For this, use the \draw command and
add the clip option. (This is not the whole picture: You can also use the \clip command and add the
draw option. Well, that is also not the whole picture: In reality, \draw is just a shorthand for \path[draw]
and \clip is a shorthand for \path[clip] and you could also say \path[draw,clip].) Here is an example:
24
\begin{tikzpicture}[scale=3]
\clip[draw] (0.5,0.5) circle (.6cm);
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\draw (3mm,0mm) arc (0:30:3mm);
\end{tikzpicture}
2.12 Parabola and Sine Path Construction
Although Karl does not need them for his picture, he is pleased to learn that there are parabola and sin and
cos path operations for adding parabolas and sine and cosine curves to the current path. For the parabola
operation, the current point will lie on the parabola as well as the point given after the parabola operation.
Consider the following example:
\tikz \draw (0,0) rectangle (1,1) (0,0) parabola (1,1);
It is also possible to place the bend somewhere else:
\tikz \draw[x=1pt,y=1pt] (0,0) parabola bend (4,16) (6,12);
The operations sin and cos add a sine or cosine curve in the interval [0, π/2] such that the previous
current point is at the start of the curve and the curve ends at the given end point. Here are two examples:
A sine curve.
A sine \tikz \draw[x=1ex,y=1ex] (0,0) sin (1.57,1); curve.
\tikz \draw[x=1.57ex,y=1ex] (0,0) sin (1,1) cos (2,0) sin (3,-1) cos (4,0)
(0,1) cos (1,0) sin (2,-1) cos (3,0) sin (4,1);
2.13 Filling and Drawing
Returning to the picture, Karl now wants the angle to be “filled” with a very light green. For this he uses
\fill instead of \draw. Here is what Karl does:
\begin{tikzpicture}[scale=3]
\clip (-0.1,-0.2) rectangle (1.1,0.75);
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\fill[green!20!white] (0,0) -- (3mm,0mm) arc (0:30:3mm) -- (0,0);
\end{tikzpicture}
The color green!20!white means 20% green and 80% white mixed together. Such color expression are
possible since pgf uses Uwe Kern’s xcolor package, see the documentation of that package for details on
color expressions.
What would have happened, if Karl had not “closed” the path using --(0,0) at the end? In this case,
the path is closed automatically, so this could have been omitted. Indeed, it would even have been better to
write the following, instead:
\fill[green!20!white] (0,0) -- (3mm,0mm) arc (0:30:3mm) -- cycle;
The --cycle causes the current path to be closed (actually the current part of the current path) by
smoothly joining the first and last point. To appreciate the difference, consider the following example:
\begin{tikzpicture}[line width=5pt]
\draw (0,0) -- (1,0) -- (1,1) -- (0,0);
\draw (2,0) -- (3,0) -- (3,1) -- cycle;
\useasboundingbox (0,1.5); % make bounding box higher
\end{tikzpicture}
25
You can also fill and draw a path at the same time using the \filldraw command. This will first draw
the path, then fill it. This may not seem too useful, but you can specify different colors to be used for filling
and for stroking. These are specified as optional arguments like this:
\begin{tikzpicture}[scale=3]
\clip (-0.1,-0.2) rectangle (1.1,0.75);
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\filldraw[fill=green!20!white, draw=green!50!black]
(0,0) -- (3mm,0mm) arc (0:30:3mm) -- cycle;
\end{tikzpicture}
2.14 Shading
Karl briefly considers the possibility of making the angle “more fancy” by shading it. Instead of filling the
with a uniform color, a smooth transition between different colors is used. For this, \shade and \shadedraw,
for shading and drawing at the same time, can be used:
\tikz \shade (0,0) rectangle (2,1) (3,0.5) circle (.5cm);
The default shading is a smooth transition from gray to white. To specify different colors, you can use
options:
\begin{tikzpicture}[rounded corners,ultra thick]
\shade[top color=yellow,bottom color=black] (0,0) rectangle +(2,1);
\shade[left color=yellow,right color=black] (3,0) rectangle +(2,1);
\shadedraw[inner color=yellow,outer color=black,draw=yellow] (6,0) rectangle +(2,1);
\shade[ball color=green] (9,.5) circle (.5cm);
\end{tikzpicture}
For Karl, the following might be appropriate:
\begin{tikzpicture}[scale=3]
\clip (-0.1,-0.2) rectangle (1.1,0.75);
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\shadedraw[left color=gray,right color=green, draw=green!50!black]
(0,0) -- (3mm,0mm) arc (0:30:3mm) -- cycle;
\end{tikzpicture}
However, he wisely decides that shadings usually only distract without adding anything to the picture.
2.15 Specifying Coordinates
Karl now wants to add the sine and cosine lines. He knows already that he can use the color= option to set
the lines’s colors. So, what is the best way to specify the coordinates?
There are different ways of specifying coordinates. The easiest way is to say something like (10pt,2cm).
This means 10pt in x-direction and 2cm in y-directions. Alternatively, you can also leave out the units as in
(1,2), which means “one times the current x-vector plus twice the current y-vector.” These vectors default
to 1cm in the x-direction and 1cm in the y-direction, respectively.
In order to specify points in polar coordinates, use the notation (30:1cm), which means 1cm in direction
30 degree. This is obviously quite useful to “get to the point (cos 30

, sin 30

) on the circle.”
You can add a single + sign in front of a coordinate or two of them as in +(1cm,0cm) or ++(0cm,2cm).
Such coordinates are interpreted differently: The first form means “1cm upwards from the previous specified
position” and the second means “2cm to the right of the previous specified position, making this the new
specified position.” For example, we can draw the sine line as follows:
26
\begin{tikzpicture}[scale=3]
\clip (-0.1,-0.2) rectangle (1.1,0.75);
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\filldraw[fill=green!20,draw=green!50!black]
(0,0) -- (3mm,0mm) arc (0:30:3mm) -- cycle;
\draw[red,very thick] (30:1cm) -- +(0,-0.5);
\end{tikzpicture}
Karl used the fact sin 30

= 1/2. However, he very much doubts that his students know this, so it would
be nice to have a way of specifying “the point straight down from (30:1cm) that lies on the x-axis.” This
is, indeed, possible using a special syntax: Karl can write (30:1cm |- 0,0). In general, the meaning of
(¸p) |- ¸q)) is “the intersection of a vertical line through p and a horizontal line through q.”
Next, let us draw the cosine line. One way would be to say (30:1cm |- 0,0) -- (0,0). Another way
is the following: we “continue” from where the sine ends:
\begin{tikzpicture}[scale=3]
\clip (-0.1,-0.2) rectangle (1.1,0.75);
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\draw (-1.5,0) -- (1.5,0);
\draw (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\filldraw[fill=green!20,draw=green!50!black] (0,0) -- (3mm,0mm) arc
(0:30:3mm) -- cycle;
\draw[red,very thick] (30:1cm) -- +(0,-0.5);
\draw[blue,very thick] (30:1cm) ++(0,-0.5) -- (0,0);
\end{tikzpicture}
Note the there is no -- between (30:1cm) and +(0,-0.5). In detail, this path is interpreted as follows:
“First, the (30:1cm) tells me to move by pen to (cos 30

, 1/2). Next, there comes another coordinate
specification, so I move my pen there without drawing anything. This new point is half a unit down from
the last position, thus it is at (cos 30

, 0). Finally, I move the pen to the origin, but this time drawing
something (because of the --).”
To appreciate the difference between + and ++ consider the following example:
\begin{tikzpicture}
\def\rectanglepath{-- ++(1cm,0cm) -- ++(0cm,1cm) -- ++(-1cm,0cm) -- cycle}
\draw (0,0) \rectanglepath;
\draw (1.5,0) \rectanglepath;
\end{tikzpicture}
By comparison, when using a single +, the coordinates are different:
\begin{tikzpicture}
\def\rectanglepath{-- +(1cm,0cm) -- +(1cm,1cm) -- +(0cm,1cm) -- cycle}
\draw (0,0) \rectanglepath;
\draw (1.5,0) \rectanglepath;
\end{tikzpicture}
Naturally, all of this could have been written more clearly and more economically like this (either with
a single of a double +):
\tikz \draw (0,0) rectangle +(1,1) (1.5,0) rectangle +(1,1);
Karl is left with the line for tan α, which seems difficult to specify using transformations and polar
coordinates. For this he needs another way of specifying coordinates: Karl can specify intersections of lines
as coordinates. The line for tan α starts at (1, 0) and goes upward to a point that is at the intersection
of a line going “up” and a line going from the origin through (30:1cm). The syntax for this point is the
following:
\draw[very thick,orange] (1,0) -- (intersection of 1,0--1,1 and 0,0--30:1cm);
In the following, two final examples of how to use relative positioning are presented. Note that the
transformation options, which are explained later, are often more useful for shifting than relative positioning.
27
\begin{tikzpicture}[scale=0.5]
\draw (0,0) -- (90:1cm) arc (90:360:1cm) arc (0:30:1cm) -- cycle;
\draw (60:5pt) -- +(30:1cm) arc (30:90:1cm) -- cycle;
\draw (3,0) +(0:1cm) -- +(72:1cm) -- +(144:1cm) -- +(216:1cm) --
+(288:1cm) -- cycle;
\end{tikzpicture}
2.16 Adding Arrow Tips
Karl now wants to add the little arrow tips at the end of the axes. He has noticed that in many plots, even
in scientific journals, these arrow tips seem to missing, presumably because the generating programs cannot
produce them. Karl thinks arrow tips belong at the end of axes. His son agrees. His students do not care
about arrow tips.
It turns out that adding arrow tips is pretty easy: Karl adds the option -> to the drawing commands for
the axes:
\begin{tikzpicture}[scale=3]
\clip (-0.1,-0.2) rectangle (1.1,1.51);
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\draw[->] (-1.5,0) -- (1.5,0);
\draw[->] (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\filldraw[fill=green!20,draw=green!50!black] (0,0) -- (3mm,0mm) arc
(0:30:3mm) -- cycle;
\draw[red,very thick] (30:1cm) -- +(0,-0.5);
\draw[blue,very thick] (30:1cm) ++(0,-0.5) -- (0,0);
\draw[orange,very thick] (1,0) -- (intersection of 1,0--1,1 and 0,0--30:1cm);
\end{tikzpicture}
If Karl had used the option <- instead of ->, arrow tips would have been put at the beginning of the
path. The option <-> puts arrow tips at both ends of the path.
There are certain restrictions to the kind of paths to which arrow tips can be added. As a rule of thumb,
you can add arrow tips only to a single open “line.” For example, you should not try to add tips to, say,
a rectangle or a circle. (You can try, but no guarantees as to what will happen now or in future versions.)
However, you can add arrow tips to curved paths and to paths that have several segments, as in the following
examples:
\begin{tikzpicture}
\draw [<->] (0,0) arc (180:30:10pt);
\draw [<->] (1,0) -- (1.5cm,10pt) -- (2cm,0pt) -- (2.5cm,10pt);
\end{tikzpicture}
Karl has a more detailed look at the arrow that TikZ puts at the end. It looks like this when he zooms
it: . The shape seems vaguely familiar and, indeed, this is exactly the end of T
E
X’s standard arrow used
in something like f : A → B.
Karl likes the arrow, especially since it is not “as thick” as the arrows offered by many other packages.
However, he expects that, sometimes, he might need to use some other kinds of arrow. To do so, Karl can
say >=¸right arrow tip kind), where ¸right arrow tip kind) is a special arrow tip specification. For example,
if Karl says >=stealth, then he tells TikZ that he would like “stealth-fighter-like” arrow tips:
\begin{tikzpicture}[>=stealth]
\draw [->] (0,0) arc (180:30:10pt);
\draw [<<-,very thick] (1,0) -- (1.5cm,10pt) -- (2cm,0pt) -- (2.5cm,10pt);
\end{tikzpicture}
Karl wonders whether such a military name for the arrow type is really necessary. He is not really
mollified when his son tells him that Microsoft’s PowerPoint uses the same name. He decides to have his
students discuss this at some point.
In addition to stealth there are several other predefined arrow tip kinds Karl can choose from, see
Section 18. Furthermore, he can define arrows types himself, if he needs new ones.
28
2.17 Scoping
Karl saw already that there are numerous graphic options that affect how paths are rendered. Often, he
would like to apply certain options to a whole set of graphic commands. For example, Karl might wish to
draw three paths using a thick pen, but would like everything else to be drawn “normally.”
If Karl wishes to set a certain graphic option for the whole picture, he can simply pass this option to
the \tikz command or to the {tikzpicture} environment (Gerda would pass the options to \tikzpicture
and Hans passes them to \starttikzpicture). However, if Karl wants to apply graphic options to a local
group, he put these commands inside a {scope} environment (Gerda uses \scope and \endscope, Hans
uses \startscope and \stopscope). This environment takes graphic options as an optional argument and
these options apply to everything inside the scope, but not to anything outside.
Here is an example:
\begin{tikzpicture}[ultra thick]
\draw (0,0) -- (0,1);
\begin{scope}[thin]
\draw (1,0) -- (1,1);
\draw (2,0) -- (2,1);
\end{scope}
\draw (3,0) -- (3,1);
\end{tikzpicture}
Scoping has another interesting effect: Any changes to the clipping area are local to the scope. Thus,
if you say \clip somewhere inside a scope, the effect of the \clip command ends at the end of the scope.
This is useful since there is no other way of “enlarging” the clipping area.
Karl has also already seen that giving options to commands like \draw apply only to that command.
In turns out that the situation is slightly more complex. First, options to a command like \draw are
not really options to the command, but they are “path options” and can be given anywhere on the
path. So, instead of \draw[thin] (0,0) -- (1,0); one can also write \draw (0,0) [thin] -- (1,0);
or \draw (0,0) -- (1,0) [thin];; all of these have the same effect. This might seem strange since in the
last case, it would appear that the thin should take effect only “after” the line from (0, 0) to (1, 0) has been
draw. However, most graphic options only apply to the whole path. Indeed, if you say both thin and thick
on the same path, the last option given will “win.”
When reading the above, Karl notices that only “most” graphic options apply to the whole path. Indeed,
all transformation options do not apply to the whole path, but only to “everything following them on the
path.” We will have a more detailed look at this in a moment. Nevertheless, all options given during a path
construction apply only to this path.
2.18 Transformations
When you specify a coordinate like (1cm,1cm), where is that coordinate placed on the page? To determine
the position, TikZ, T
E
X, and pdf or PostScript all apply certain transformations to the given coordinate in
order to determine the finally position on the page.
TikZ provides numerous options that allow you to transform coordinates in pgf’s private coordinate
system. For example, the xshift option allows you to shift all subsequent points by a certain amount:
\tikz \draw (0,0) -- (0,0.5) [xshift=2pt] (0,0) -- (0,0.5);
It is important to note that you can change transformation “in the middle of a path,” a feature that is
not supported by pdf or PostScript. The reason is that pgf keeps track of its own transformation matrix.
Here is a more complicated example:
\begin{tikzpicture}[even odd rule,rounded corners=2pt,x=10pt,y=10pt]
\filldraw[fill=examplefill] (0,0) rectangle (1,1)
[xshift=5pt,yshift=5pt] (0,0) rectangle (1,1)
[rotate=30] (-1,-1) rectangle (2,2);
\end{tikzpicture}
The most useful transformations are xshift and yshift for shifting, shift for shifting to a given point
as in shift={(1,0)} or shift={+(0,0)} (the braces are necessary so that T
E
X does not mistake the comma
for separating options), rotate for rotating by a certain angle (there is also a rotate around for rotating
around a given point), scale for scaling by a certain factor, xscale and yscale for scaling only in the x-
or y-direction (xscale=-1 is a flip), and xslant and yslant for slanting. If these transformation and those
29
that I have not mentioned are not sufficient, the cm option allows you to apply an arbitrary transformation
matrix. Karl’s students, by the way, do not know what a transformation matrix is.
2.19 Repeating Things: For-Loops
Karl’s next aim is to add little ticks on the axes at positions −1, −1/2, 1/2, and 1. For this, it would be
nice to use some kind of “loop,” especially since he wishes to do the same thing at each of these positions.
There are different packages for doing this. L
A
T
E
X has its own internal command for this, pstricks comes
along with the powerful \mulitdo command. All of these can be used together with pgf and TikZ, so if
you are familiar with them, feel free to use them. pgf introduces yet another command, called \foreach,
which I introduced since I could never remember the syntax of the other packages. \foreach is defined in
the package pgffor and can be used independently of pgf. TikZ includes it automatically.
In its basic form, the \foreach command is easy to use:
x = 1, x = 2, x = 3, \foreach \x in {1,2,3} {$x =\x$, }
The general syntax is \foreach ¸variable) in {¸list of values)} ¸commands). Inside the ¸commands),
the ¸variable) will be assigned to the different values. If the ¸commands) do not start with a brace, everything
up to the next semicolon is used as ¸commands).
For Karl and the ticks on the axes, he could use the following code:
\begin{tikzpicture}[scale=3]
\clip (-0.1,-0.2) rectangle (1.1,1.51);
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\filldraw[fill=green!20,draw=green!50!black] (0,0) -- (3mm,0mm) arc
(0:30:3mm) -- cycle;
\draw[->] (-1.5,0) -- (1.5,0);
\draw[->] (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\foreach \x in {-1cm,-0.5cm,1cm}
\draw (\x,-1pt) -- (\x,1pt);
\foreach \y in {-1cm,-0.5cm,0.5cm,1cm}
\draw (-1pt,\y) -- (1pt,\y);
\end{tikzpicture}
As a matter of fact, there are many different ways of creating the ticks. For example, Karl could have
put the \draw ...; inside curly braces. He could also have used, say,
\foreach \x in {-1,-0.5,1}
\draw[xshift=\x cm] (0pt,-1pt) -- (0pt,1pt);
Karl is curious what would happen in a more complicated situation where there are, say, 20 ticks. It
seems bothersome to explicitly mention all these numbers in the set for \foreach. Indeed, it is possible to
use ... inside the \foreach statement to iterate over a large number of values (which must, however, be
dimensionless real numbers) as in the following example:
\tikz \foreach \x in {1,...,10}
\draw (\x,0) circle (0.4cm);
If you provide two numbers before the ..., the \foreach statement will use their difference for the
stepping:
\tikz \foreach \x in {-1,-0.5,...,1}
\draw (\x cm,-1pt) -- (\x cm,1pt);
We can also nest loops to create interesting effects:
30
1,1
1,2
1,3
1,4
1,5
2,1
2,2
2,3
2,4
2,5
3,1
3,2
3,3
3,4
3,5
4,1
4,2
4,3
4,4
4,5
5,1
5,2
5,3
5,4
5,5
7,1
7,2
7,3
7,4
7,5
8,1
8,2
8,3
8,4
8,5
9,1
9,2
9,3
9,4
9,5
10,1
10,2
10,3
10,4
10,5
11,1
11,2
11,3
11,4
11,5
12,1
12,2
12,3
12,4
12,5
\begin{tikzpicture}
\foreach \x in {1,2,...,5,7,8,...,12}
\foreach \y in {1,...,5}
{
\draw (\x,\y) +(-.5,-.5) rectangle ++(.5,.5);
\draw (\x,\y) node{\x,\y};
}
\end{tikzpicture}
The \foreach statement can do even trickier stuff, but the above gives the idea.
2.20 Adding Text
Karl is, by now, quite satisfied with the picture. However, the most important parts, namely the labels, are
still missing!
TikZ offers an easy-to-use and powerful system for adding text and, more generally, complex shapes
to a picture at specific positions. The basic idea is the following: When TikZ is constructing a path and
encounters the keyword node in the middle of a path, it reads a node specification. The keyword node is
typically followed by some options and then some text between curly braces. This text is put inside a normal
T
E
X box (if the node specification directly follows a coordinate, which is usually the case, TikZ is able to
perform some magic so that it is even possible to use verbatim text inside the boxes) and then placed at the
current position, that is, at the last specified position (possibly shifted a bit, according to the given options).
However, all nodes are drawn only after the path has been completely drawn/filled/shaded/clipped/whatever.
Text at node 1
Text at node 2
\begin{tikzpicture}
\draw (0,0) rectangle (2,2);
\draw (0.5,0.5) node [fill=examplefill]
{Text at \verb!node 1!}
-- (1.5,1.5) node {Text at \verb!node 2!};
\end{tikzpicture}
Obviously, Karl would not only like to place nodes on the last specified position, but also to the left
or the right of these positions. For this, every node object that you put in your picture is equipped with
several anchors. For example, the north anchor is in the middle at the upper end of the shape, the south
anchor is at the bottom and the north east anchor is in the upper right corner. When you given the option
anchor=north, the text will be placed such that this northern anchor will lie on the current position and
the text is, thus, below the current position. Karl uses this to draw the ticks as follows:
31
−1 −0.5 1
−1
−0.5
0.5
1
\begin{tikzpicture}[scale=3]
\clip (-0.6,-0.2) rectangle (0.6,1.51);
\draw[step=.5cm,style=help lines] (-1.4,-1.4) grid (1.4,1.4);
\filldraw[fill=green!20,draw=green!50!black]
(0,0) -- (3mm,0mm) arc (0:30:3mm) -- cycle;
\draw[->] (-1.5,0) -- (1.5,0); \draw[->] (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\foreach \x in {-1,-0.5,1}
\draw (\x cm,1pt) -- (\x cm,-1pt) node[anchor=north] {$\x$};
\foreach \y in {-1,-0.5,0.5,1}
\draw (1pt,\y cm) -- (-1pt,\y cm) node[anchor=east] {$\y$};
\end{tikzpicture}
This is quite nice, already. Using these anchors, Karl can now add most of the other text elements.
However, Karl thinks that, though “correct,” it is quite counter-intuitive that in order to place something
below a given point, he has to use the north anchor. For this reason, there is an option called below, which
does the same as anchor=north. Similarly, above right does the same as anchor=south east. In addition,
below takes an optional dimension argument. If given, the shape will additionally be shifted downwards by
the given amount. So, below=1pt can be used to put a text label below some point and, additionally shift
it 1pt downwards.
Karl is not quite satisfied with the ticks. He would like to have 1/2 or
1
2
shown instead of 0.5, partly to
show off the nice capabilities of T
E
X and TikZ, partly because for positions like 1/3 or π it is certainly very
much preferable to have the “mathematical” tick there instead of just the “numeric” tick. His students, on
the other hand, prefer 0.5 over 1/2 since they are not too fond of fractions in general.
Karl now faces a problem: For the \foreach statement, the position \x should still be given as 0.5 since
TikZ will not know where \frac{1}{2} is supposed to be. On the other hand, the typeset text should really
be \frac{1}{2}. To solve this problem, \foreach offers a special syntax: Instead of having one variable \x,
Karl can specify two (or even more) variables separated by a slash as in \x / \xtext. Then, the elements
in the set over which \foreach iterates must also be of the form ¸first)/¸second). In each iteration, \x will
be set to ¸first) and \xtext will be set to ¸second). If no ¸second) is given, the ¸first) will be used again.
So, here is the new code for the ticks:
−1

1
2
1
−1

1
2
1
2
1
\begin{tikzpicture}[scale=3]
\clip (-0.6,-0.2) rectangle (0.6,1.51);
\draw[step=.5cm,style=help lines] (-1.4,-1.4) grid (1.4,1.4);
\filldraw[fill=green!20,draw=green!50!black]
(0,0) -- (3mm,0mm) arc (0:30:3mm) -- cycle;
\draw[->] (-1.5,0) -- (1.5,0); \draw[->] (0,-1.5) -- (0,1.5);
\draw (0,0) circle (1cm);
\foreach \x/\xtext in {-1, -0.5/-\frac{1}{2}, 1}
\draw (\x cm,1pt) -- (\x cm,-1pt) node[anchor=north] {$\xtext$};
\foreach \y/\ytext in {-1, -0.5/-\frac{1}{2}, 0.5/\frac{1}{2}, 1}
\draw (1pt,\y cm) -- (-1pt,\y cm) node[anchor=east] {$\ytext$};
\end{tikzpicture}
Karl is quite pleased with the result, but his son points out that this is still not perfectly satisfactory:
The grid and the circle interfere with the numbers and decrease their legibility. Karl is not very concerned
by this (his students do not even notice), but his son insists that there is an easy solution: Karl can add the
[fill=white] option to fill out the background of the text shape with a white color.
The next thing Karl wants to do is to add the labels like sin α. For this, he would like to place a label
“in the middle of line.” To do so, instead of specifying the label node {$\sin\alpha$} directly after one
of the endpoints of the line (which would place the label at that endpoint), Karl can give the label directly
after the --, before the coordinate. By default, this places the label in the middle of the line, but the pos=
options can be used to modify this. Also, options like near start and near end can be used to modify this
position:
32
sin α
cos α
tan α =
sin α
cos α
−1

1
2
1
−1

1
2
1
2
1
\begin{tikzpicture}[scale=3]
\clip (-2,-0.2) rectangle (2,0.8);
\draw[step=.5cm,gray,very thin] (-1.4,-1.4) grid (1.4,1.4);
\filldraw[fill=green!20,draw=green!50!black] (0,0) -- (3mm,0mm) arc
(0:30:3mm) -- cycle;
\draw[->] (-1.5,0) -- (1.5,0) coordinate (x axis);
\draw[->] (0,-1.5) -- (0,1.5) coordinate (y axis);
\draw (0,0) circle (1cm);
\draw[very thick,red]
(30:1cm) -- node[left=1pt,fill=white] {$\sin \alpha$} (30:1cm |- x axis);
\draw[very thick,blue]
(30:1cm |- x axis) -- node[below=2pt,fill=white] {$\cos \alpha$} (0,0);
\draw[very thick,orange] (1,0) -- node [right=1pt,fill=white]
{$\displaystyle \tan \alpha \color{black}=
\frac{{\color{red}\sin \alpha}}{\color{blue}\cos \alpha}$}
(intersection of 0,0--30:1cm and 1,0--1,1) coordinate (t);
\draw (0,0) -- (t);
\foreach \x/\xtext in {-1, -0.5/-\frac{1}{2}, 1}
\draw (\x cm,1pt) -- (\x cm,-1pt) node[anchor=north,fill=white] {$\xtext$};
\foreach \y/\ytext in {-1, -0.5/-\frac{1}{2}, 0.5/\frac{1}{2}, 1}
\draw (1pt,\y cm) -- (-1pt,\y cm) node[anchor=east,fill=white] {$\ytext$};
\end{tikzpicture}
You can also position labels on curves and, by adding the sloped option, have them rotated such that
they match the line’s slope. Here is an example:
near start
midway
v
e
ry
n
e
a
r
e
n
d
\begin{tikzpicture}
\draw (0,0) .. controls (6,1) and (9,1) ..
node[near start,sloped,above] {near start}
node {midway}
node[very near end,sloped,below] {very near end} (12,0);
\end{tikzpicture}
It remains to draw the explanatory text at the right of the picture. The main difficulty here lies in
limiting the width of the text “label,” which is quite long, so that line breaking is used. Fortunately, Karl
can use the option text width=6cm to get the desired effect. So, here is the full code:
33
\begin{tikzpicture}[scale=3,cap=round]
% Local definitions
\def\costhirty{0.8660256}
% Colors
\colorlet{anglecolor}{green!50!black}
\colorlet{sincolor}{red}
\colorlet{tancolor}{orange!80!black}
\colorlet{coscolor}{blue}
% Styles
\tikzstyle{axes}=[]
\tikzstyle{important line}=[very thick]
\tikzstyle{information text}=[rounded corners,fill=red!10,inner sep=1ex]
% The graphic
\draw[style=help lines,step=0.5cm] (-1.4,-1.4) grid (1.4,1.4);
\draw (0,0) circle (1cm);
\begin{scope}[style=axes]
\draw[->] (-1.5,0) -- (1.5,0) node[right] {$x$} coordinate(x axis);
\draw[->] (0,-1.5) -- (0,1.5) node[above] {$y$} coordinate(y axis);
\foreach \x/\xtext in {-1, -.5/-\frac{1}{2}, 1}
\draw[xshift=\x cm] (0pt,1pt) -- (0pt,-1pt) node[below,fill=white] {$\xtext$};
\foreach \y/\ytext in {-1, -.5/-\frac{1}{2}, .5/\frac{1}{2}, 1}
\draw[yshift=\y cm] (1pt,0pt) -- (-1pt,0pt) node[left,fill=white] {$\ytext$};
\end{scope}
\filldraw[fill=green!20,draw=anglecolor] (0,0) -- (3mm,0pt) arc(0:30:3mm);
\draw (15:2mm) node[anglecolor] {$\alpha$};
\draw[style=important line,sincolor]
(30:1cm) -- node[left=1pt,fill=white] {$\sin \alpha$} (30:1cm |- x axis);
\draw[style=important line,coscolor]
(30:1cm |- x axis) -- node[below=2pt,fill=white] {$\cos \alpha$} (0,0);
\draw[style=important line,tancolor] (1,0) -- node[right=1pt,fill=white] {
$\displaystyle \tan \alpha \color{black}=
\frac{{\color{sincolor}\sin \alpha}}{\color{coscolor}\cos \alpha}$}
(intersection of 0,0--30:1cm and 1,0--1,1) coordinate (t);
\draw (0,0) -- (t);
\draw[xshift=1.85cm]
node[right,text width=6cm,style=information text]
{
The {\color{anglecolor} angle $\alpha$} is $30^\circ$ in the
example ($\pi/6$ in radians). The {\color{sincolor}sine of
$\alpha$}, which is the height of the red line, is
\[
{\color{sincolor} \sin \alpha} = 1/2.
\]
By the Theorem of Pythagoras ...
};
\end{tikzpicture}
34
3 Tutorial: A Petri-Net for Hagen
In this second tutorial we explore the node mechanism of TikZ and pgf.
Hagen must give a talk tomorrow about his favorite formalism for distributed systems: Petri nets!
Hagen used to give his talks using a blackboard and everyone seemed to be perfectly concent with this.
Unfortunately, his audience has been spoiled recently with fancy projector-based presentations and there
seems to be a certain amount of peer pressure that this Petri nets should also be drawn using a graphic
program. One of the professors at his institutes recommends TikZ for this and Hagen decides to give it a
try.
3.1 Problem Statement
For his talk, Hagen wishes to create a graphic that demonstrates how a net with place capacities can be
simulated by a net without capacities. The graphic should look like this, ideally:
s ≤ 3
2
2
s ¯ s
2
2
replacement of
the capacity
by two places
3.2 Setting up the Environment
For the picture Hagen will need to load the TikZ package as did Karl in the previous tutorial. However,
Hagen will also need to load some additional library packages that Karl did not need. These library packages
contain additional definitions like extra arrow tips that are typically not needed in a picture and that need
to be loaded explicitly.
Hagen will need to load three libraries: The arrow tip library for the special arrow tip used in the graphic,
the snake library with the “snaking line” in the middle, and the background library for the two rectangular
areas that are behind the two main parts of the picture.
3.2.1 Setting up the Environment in L
A
T
E
X
When using L
A
T
E
X use:
\documentclass{article} % say
\usepackage{tikz}
\usetikzlibrary{arrows,snakes,backgrounds}
\begin{document}
\begin{tikzpicture}
\draw (0,0) -- (1,1);
\end{tikzpicture}
\end{document}
3.2.2 Setting up the Environment in Plain T
E
X
When using plain T
E
X use:
35
%% Plain TeX file
\input tikz.tex
\usetikzlibrary{arrows,snakes,backgrounds}
\baselineskip=12pt
\hsize=6.3truein
\vsize=8.7truein
\tikzpicture
\draw (0,0) -- (1,1);
\endtikzpicture
\bye
3.2.3 Setting up the Environment in ConT
E
Xt
When using ConT
E
X use:
%% ConTeXt file
\usemodule[tikz]
\usetikzlibrary[arrows,snakes,backgrounds]
\starttext
\starttikzpicture
\draw (0,0) -- (1,1);
\stoptikzpicture
\startext
3.3 Introduction to Nodes
In principle, we already know how to create the graphics that Hagen desires (except perhaps for the snaked
line, we will come to that): We start with big light gray rectangle and then add lots of circles and small
rectangle, plus some arrows.
However, this approach has numerous disadvantages: First, it is hard to change anything at a later stage.
For example, if we decide to add more places to the Petri nets (the circles are called places in Petri net
theory), all of the coordinates change and we need to recalculate everything. Second, it is hard to read the
code for the Petri net as it just a long and complicated list of coordinates and drawing commands – the
underlying structure of the Petri net is lost.
Fortunately, TikZ offers a powerful mechanism for avoiding the above problems: nodes. We already came
across nodes in the previous tutorial, where we used them to add labels to Karl’s graphic. In the present
tutorial we will see that nodes are much more powerful.
A node is a small part of a picture. When a node is created, you provide a position where the node
should be drawn and a shape. A node of shape circle will be drawn as a circle, a node of shape rectangle
as a rectangle, and so on. A node may also contain same text, which is why Karl used nodes to show text.
Finally, a node can get a name for later reference.
In Hagen’s picture we will use nodes for the places and for the transitions of the Petri net (the places are
the circles, the transitions are the rectangles). Let us start with the upper half of the left Petir net. In this
upper half we have three places and two transitions. Instead of drawing three circles and two rectangles, we
use three nodes of shape circle and two nodes of shape rectangle.
\begin{tikzpicture}
\path ( 0,2) node [shape=circle,draw] {}
( 0,1) node [shape=circle,draw] {}
( 0,0) node [shape=circle,draw] {}
( 1,1) node [shape=rectangle,draw] {}
(-1,1) node [shape=rectangle,draw] {};
\end{tikzpicture}
Hagen notes that this does not quite look like the final picture, but it seems like a good first step.
Let us have a more detailed look at the code. The whole picture consists of a single path. Ignoring the
node operations there is not much going on in this path: It is just a sequence of coordinates with nothing
“happening” between them. Indeed, even if something were to happen like a line-to or a curve-to, the \path
command would not “do” anything with the resulting path. So, all the magic must be in the node commands.
In the previous tutorial we learned that a node will add a piece of text at the last coordinate. Thus,
each of the five nodes is added at a different position. In the above code, this text is empty (because of the
empty {}). So, why do we see anything at all? The answer is the draw option for the node operation: It
causes the “shape around the text” to be drawn.
36
So, the code (0,2) node [shape=circle,draw] {} means the following: “In the main path, add a
move-to to the coordinate (0,2). Then, temporarily suspend the construction of the main path while the
node is build. This node will be a circle around an empty text. This circle is to be drawn, but not filled
or otherwise used. Once this whole node is constructed, it is saved until after the main path is finished.
Then, it is drawn.” Then following (0,1) node [shape=circle,draw] {} then has the following effect:
“Continue the main path with a move-to to (0,1). Then construct a node at this position also. This node
is also shown after the main path is finished.” And so on.
3.4 Placing Nodes Using the At Syntax
Hagen now understands how the node operation adds nodes to the path, but it seems a bit silly to create a
path using the \path operation, consisting of numerous superfluous move-to operations, only to place nodes.
He is pleased to learn that there are ways to add nodes in a more sensible manner.
First, the node operation allows one to add at (¸coordinate)) in order to directly specify where the node
should be placed, sidestepping the rule that nodes are placed on the last coordinate. Hagen can then write
the following:
\begin{tikzpicture}
\path node at ( 0,2) [shape=circle,draw] {}
node at ( 0,1) [shape=circle,draw] {}
node at ( 0,0) [shape=circle,draw] {}
node at ( 1,1) [shape=rectangle,draw] {}
node at (-1,1) [shape=rectangle,draw] {};
\end{tikzpicture}
Now Hagen is still left with a single empty path, but at least the path no longer contains strange move-
tos. It turns out that this can be improved further: The \node command is an abbreviation for \path node,
which allows Hagen to write:
\begin{tikzpicture}
\node at ( 0,2) [circle,draw] {};
\node at ( 0,1) [circle,draw] {};
\node at ( 0,0) [circle,draw] {};
\node at ( 1,1) [rectangle,draw] {};
\node at (-1,1) [rectangle,draw] {};
\end{tikzpicture}
Hagen likes this syntax much better than the previous one. Note that Hagen has also omitted the shape=
since, like color=, TikZ allows you to omit the shape= if there is no confusion.
3.5 Using Styles
Feeling adventurous, Hagen tries to make the nodes look nicer. In the final picture, the circles and rectangle
should be filled with different colors, resulting in the following code:
\begin{tikzpicture}[thick]
\node at ( 0,2) [circle,draw=blue!50,fill=blue!20] {};
\node at ( 0,1) [circle,draw=blue!50,fill=blue!20] {};
\node at ( 0,0) [circle,draw=blue!50,fill=blue!20] {};
\node at ( 1,1) [rectangle,draw=black!50,fill=black!20] {};
\node at (-1,1) [rectangle,draw=black!50,fill=black!20] {};
\end{tikzpicture}
While this looks nicer in the picture, the code starts to get a bit ugly. Ideally, we would like our code
to transport the message “there are three places and two transitions” and not so much which filling colors
should be used.
To solve this problem, Hagen uses styles. He defines a style for places and another style for transitions:
\tikzstyle{place}=[circle,draw=blue!50,fill=blue!20,thick]
\tikzstyle{transition}=[rectangle,draw=black!50,fill=black!20,thick]
\begin{tikzpicture}
\node at ( 0,2) [place] {};
\node at ( 0,1) [place] {};
\node at ( 0,0) [place] {};
\node at ( 1,1) [transition] {};
\node at (-1,1) [transition] {};
\end{tikzpicture}
37
3.6 Node Size
Before Hagen starts naming and connecting the nodes, let us first make sure that the nodes get their final
appearance. They are still too small. Indeed, Hagen wonders why they have any size at all, after all, the
text is empty. The reason is than TikZ automatically adds some space around the text. The amount is set
using the option inner sep. So, to increase the size of the nodes, Hagen could write:
\tikzstyle{place}=[circle,draw=blue!50,fill=blue!20,thick]
\tikzstyle{transition}=[rectangle,draw=black!50,fill=black!20,thick]
\begin{tikzpicture}[inner sep=2mm]
\node at ( 0,2) [place] {};
\node at ( 0,1) [place] {};
\node at ( 0,0) [place] {};
\node at ( 1,1) [transition] {};
\node at (-1,1) [transition] {};
\end{tikzpicture}
However, this is not really the best way to achieve the desired effect. It is much better to use the
minimum size option instead. This option allows Hagen to specify a minimum size that the node should
have. If the nodes actually needs to be bigger because of a longer text, it will be larger, but if the text
is empty, then the node will have minimum size. This option is also useful to ensure that several nodes
containing different amounts of text have the same size. The options minimum height and minimum width
allow you to specify the minimum height and width independently.
So, what Hagen needs to do is to provide minimum size for the nodes. To be on the safe side, he also
sets inner sep=0pt. This ensures that the nodes will really have size minimum size and not, for very small
minimum sizes, the minimal size necessary to encompass the automatically added space.
\tikzstyle{place}=[circle,draw=blue!50,fill=blue!20,thick,
inner sep=0pt,minimum size=6mm]
\tikzstyle{transition}=[rectangle,draw=black!50,fill=black!20,thick,
inner sep=0pt,minimum size=4mm]
\begin{tikzpicture}
\node at ( 0,2) [place] {};
\node at ( 0,1) [place] {};
\node at ( 0,0) [place] {};
\node at ( 1,1) [transition] {};
\node at (-1,1) [transition] {};
\end{tikzpicture}
3.7 Naming Nodes
Hagen’s next aim is to connect the nodes using arrows. This seems like a tricky business since the arrows
should not start in the middle of the nodes, but somewhere on the border and Hagen would very much like
to avoid computing these positions by hand.
Fortunately, pgf will perform all the necessary calculations for him. However, he first has to assign
names to the nodes so that he can reference them later on.
There are two ways to name a node. The first is the use the name= option. The second method is to
write the desired name in parentheses after the node operation. Hagen thinks that this second method seems
strange, but he will soon change his opinion.
% ... setup styles
\begin{tikzpicture}
\node (waiting 1) at ( 0,2) [place] {};
\node (critical 1) at ( 0,1) [place] {};
\node (semaphore) at ( 0,0) [place] {};
\node (leave critical) at ( 1,1) [transition] {};
\node (enter critical) at (-1,1) [transition] {};
\end{tikzpicture}
Hagen is pleased to note that the names help in understanding the code. Names for nodes can be
pretty arbitrary, but they should not contain commas, periods, parentheses, colons, and some other special
characters. However, they can contain underscores and hyphens.
The syntax for the node operation is quite liberal with respect to the order in which node names, the at
specifier, and the options must come. Indeed, you can even have multiple option blocks between the node
and the text in curly braces, they accumulate. You can rearrange them arbitrarily and perhaps the following
might be preferable:
38
\begin{tikzpicture}
\node[place] (waiting 1) at ( 0,2) {};
\node[place] (critical 1) at ( 0,1) {};
\node[place] (semaphore) at ( 0,0) {};
\node[transition] (leave critical) at ( 1,1) {};
\node[transition] (enter critical) at (-1,1) {};
\end{tikzpicture}
3.8 Placing Nodes Using Relative Placement
Although Hagen still wishes to connect the nodes, he first wishes to address another problem again: The
placement of the nodes. Although he likes the at syntax, in this particular case he would prefer placing the
nodes “relative to each other.” So, Hagen would like to say that the critical 1 node should be below the
waiting 1 node, wherever the waiting 1 node might be. There are different ways of achieving this, but
the nicest one in Hagen’s case is the below of option:
\begin{tikzpicture}
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical] {};
\node[transition] (leave critical) [right of=critical] {};
\node[transition] (enter critical) [left of=critical] {};
\end{tikzpicture}
The below of and similar options setup the position of the node in such a manner that it is placed at
the distance node distance in the specified direction of the given direction. The node distance is the
distance between the centers of the nodes, not between the borders.
Even though the above code has the same effect the earlier code, Hagen can pass it to his colleagues how
will be able to just read and understand it, perhaps without even having to see the picture.
3.9 Adding Labels Next to Nodes
Before we have a look at how Hagen can connect the nodes, let us add the capacity “s ≤ 3” to the bottom
node. For this, two approaches are possible:
1. Hagen can just add a new node above the north anchor of the semaphore node.
s ≤ 3
\begin{tikzpicture}
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical] {};
\node[transition] (leave critical) [right of=critical] {};
\node[transition] (enter critical) [left of=critical] {};
\node [red,above] at (semaphore.north) {$s\le 3$};
\end{tikzpicture}
This is a general approach that will “always work.”
2. Hagen can use the special label option. This option is given to a node and it causes another node
to be added next to the node where the option is given. Here is the idea: When we construct the
semaphore node, we wish to indicate that we want another node with the capacity above it. For this,
we use the option label=above:$s\le 3$. This option is interpreted as follows: We want a node
above the semaphore node and this node should read “s ≤ 3.” Instead of above we could also use
things like below left before the colon or a number like 60.
s ≤ 3
\begin{tikzpicture}
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical,
label=above:$s\le3$] {};
\node[transition] (leave critical) [right of=critical] {};
\node[transition] (enter critical) [left of=critical] {};
\end{tikzpicture}
39
It is also possible to give multiple label options, this causes multiple labels to be drawn.
my circle
60

−90

\tikz
\node [circle,draw,label=60:$60^\circ$,label=below:$-90^\circ$] {my circle};
Hagen is not fully satisfied with the label option since the label is not red. To achieve this, has has
two options: First, he can redefine the every label style. Second, he can add options to the label’s
node. These options are given following the label=, so he would write label=[red]above:$s\le3$.
However, this does not quite work since T
E
X thinks that the ] closes the whole option list of the
semaphore node. So, Hagen has to add braces and writes label={[red]above:$s\le3$}. Since this
looks a bit ugly, Hagen decides to redefine the every label style.
s ≤ 3
\begin{tikzpicture}
\tikzstyle{every label}=[red]
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical,
label=above:$s\le3$] {};
\node[transition] (leave critical) [right of=critical] {};
\node[transition] (enter critical) [left of=critical] {};
\end{tikzpicture}
3.10 Connecting Nodes
It is now high time to connect the nodes. Let us start with something simple, namely with the straight line
from enter critical to critical. We want this line to start at the right side of enter critical and to
end at the left side of critical. For this, we can use the anchors of the nodes. Every node defines a whole
bunch of anchors that lie on its border or inside it. For example, the center anchor is at the center of the
node, the west anchor is on the left of the node, and so on. To access the coordinate of a node, we use a
coordinate that contains the node’s name followed by a dot, followed by the anchor’s name:
\begin{tikzpicture}
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical] {};
\node[transition] (leave critical) [right of=critical] {};
\node[transition] (enter critical) [left of=critical] {};
\draw [->] (critical.west) -- (enter critical.east);
\end{tikzpicture}
Next, let us tackle the curve from waiting to enter critical. This can be specified using curves and
controls:
\begin{tikzpicture}
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical] {};
\node[transition] (leave critical) [right of=critical] {};
\node[transition] (enter critical) [left of=critical] {};
\draw [->] (enter critical.east) -- (critical.west);
\draw [->] (waiting.west) .. controls +(left:5mm) and +(up:5mm)
.. (enter critical.north);
\end{tikzpicture}
Hagen sees how he can now add all his edges, but the whole process seems a but awkward and not very
flexible. Again, the code seems to obscure the structure of the graphic rather than showing it.
So, let us start improving the code for the edges. First, Hagen can leave out the anchors:
40
\begin{tikzpicture}
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical] {};
\node[transition] (leave critical) [right of=critical] {};
\node[transition] (enter critical) [left of=critical] {};
\draw [->] (enter critical) -- (critical);
\draw [->] (waiting) .. controls +(left:8mm) and +(up:8mm)
.. (enter critical);
\end{tikzpicture}
Hagen is a bit surprised that this works. After all, how did TikZ know that the line from enter critical
to critical should actually start on the borders? Whenever TikZ encounters a whole node name as a
“coordinate,” it tries to “be smart” about the anchor that it should choose for this node. Depending on
what happens next, TikZ will choose an anchor that lies on the border of the node on a line to the next
coordinate or control point. The exact rules are a bit complex, but the chosen point will usually be correct
– and when it is not, Hagen can still specify the desired anchor by hand.
Hagen would now like to simplify the curve operation somehow. It turns out that this can be accomplished
using a special path operation: the to operation. This operation takes many options (you can even define
new ones yourself). One pair of option is useful for Hagen: The pair in and out. These options take angles
at which a curve should leave or reach the start or target coordinates. Without these options, a straight line
is drawn:
\begin{tikzpicture}
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical] {};
\node[transition] (leave critical) [right of=critical] {};
\node[transition] (enter critical) [left of=critical] {};
\draw [->] (enter critical) to (critical);
\draw [->] (waiting) to [out=180,in=90] (enter critical);
\end{tikzpicture}
There is another option for the to operation, that is even better suited to Hagen’s problem: The
bend right option. This option also takes an angle, but this angle only specifies the angle by which
the curve is bend to the right:
\begin{tikzpicture}
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical] {};
\node[transition] (leave critical) [right of=critical] {};
\node[transition] (enter critical) [left of=critical] {};
\draw [->] (enter critical) to (critical);
\draw [->] (waiting) to [bend right=45] (enter critical);
\draw [->] (enter critical) to [bend right=45] (semaphore);
\end{tikzpicture}
It is now time for Hagen to learn about yet another way of specifying edges: Using the edge path
operation. This operation is very similar to the to operation, but there is one important difference: Like a
node the edge generated by the edge operation is not part of the main path, but is added only later. This
may not seem very important, but is has some nice consequences. For example, every edge can have its own
arrow tips and its own color and so one and, still, all the edges can be given on the same path. This allows
Hagen to write the following:
\begin{tikzpicture}
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical] {};
\node[transition] (leave critical) [right of=critical] {};
\node[transition] (enter critical) [left of=critical] {}
edge [->] (critical)
edge [<-,bend left=45] (waiting)
edge [->,bend right=45] (semaphore);
\end{tikzpicture}
Each edge caused a new path to be constructed, consisting of a to between the node enter critical
and the node following the edge command.
The finishing touch is to introduce two styles pre and post and to use the bend angle=45 option to set
the bend angle once and for all:
41
% Styles place and transition as before
\tikzstyle{pre}=[<-,shorten <=1pt,>=stealth’,semithick]
\tikzstyle{post}=[->,shorten >=1pt,>=stealth’,semithick]
\begin{tikzpicture}[bend angle=45]
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical] {};
\node[transition] (leave critical) [right of=critical] {}
edge [pre] (critical)
edge [post,bend right] (waiting)
edge [pre, bend left] (semaphore);
\node[transition] (enter critical) [left of=critical] {}
edge [post] (critical)
edge [pre, bend left] (waiting)
edge [post,bend right] (semaphore);
\end{tikzpicture}
3.11 Adding Labels Next to Lines
The next thing that Hagen needs to add is the “2” at the arcs. For this Hagen can use TikZ’s automatic
node placement: By adding the option auto, TikZ will position nodes on curves and lines in such a way that
they are not on the curve but next to it. Adding swap will mirror the label with respect to the line. Here is
a general example:
0

120

240

1
1’
2 2’
3
3’
\begin{tikzpicture}[auto,bend right]
\node (a) at (0:1) {$0^\circ$};
\node (b) at (120:1) {$120^\circ$};
\node (c) at (240:1) {$240^\circ$};
\draw (a) to node {1} node [swap] {1’} (b)
(b) to node {2} node [swap] {2’} (c)
(c) to node {3} node [swap] {3’} (a);
\end{tikzpicture}
What is happening here? The nodes are given somehow inside the to operation! When this is done, the
node is placed on the middle of the curve or line created by the to operation. The auto option then causes
the node to be moved in such a way that it does not lie on the curve, but next to it. In the example we
provide even two nodes on each to operation.
For Hagen that auto option is not really necessary since the two “2” labels could also easily be placed
“by hand.” However, in a complicated plot with numerous edges automatic placement can be a blessing.
2
% Styles as before
\begin{tikzpicture}[bend angle=45]
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical] {};
\node[transition] (leave critical) [right of=critical] {}
edge [pre] (critical)
edge [post,bend right] node[auto,swap] {2} (waiting)
edge [pre, bend left] (semaphore);
\node[transition] (enter critical) [left of=critical] {}
edge [post] (critical)
edge [pre, bend left] (waiting)
edge [post,bend right] (semaphore);
\end{tikzpicture}
3.12 Adding the Snaked Line and Multi-Line Text
With the node mechanism Hagen can now easily create the two Petri nets. What he is unsure of is how he
can create the snaked line between the nets.
For this he can use a snake. Snakes a called thus since the most basic form of a snake looks exactly like
a snake. However, and repeating pattern can be used as a snake like bumps or a saw or even much more
complicated stuff.
To draw the snake, Hagen only needs to set the snake=snake option on the path. This causes all straight
lines of the path to be replaced by snakes. It is also possible to use snakes only in certain parts of a path,
but Hagen will not need this.
42
\begin{tikzpicture}
\draw [->,snake=snake] (0,0) -- (2,0);
\end{tikzpicture}
Well, that does not look quite right, yet. The problem is that the snake happens to end exactly at the
position where the arrow begins. Fortunately, there is an option that helps here. Also, the snake should be
a bit smaller, which can be influenced by even more options.
\begin{tikzpicture}
\draw [->,snake=snake,
segment amplitude=.4mm,
segment length=2mm,
line after snake=1mm] (0,0) -- (3,0);
\end{tikzpicture}
Now Hagen needs to add the text above the snake. This text is a bit challenging since it is a multi-line
text. To typeset such text, Hagen needs to specify a width for the text and he needs to specify that the text
should be centered.
replacement of
the capacity
by two places
\begin{tikzpicture}
\draw [->,snake=snake,
segment amplitude=.4mm,
segment length=2mm,
line after snake=1mm] (0,0) -- (3,0)
node [above,text width=3cm,text centered,midway]
{
replacement of the \textcolor{red}{capacity} by
\textcolor{red}{two places}
};
\end{tikzpicture}
3.13 Using Layers: The Background Rectangles
Hagen still needs to add the background rectangles. These are a bit tricky: Hagen would like to draw the
rectangles after the Petri nets are finished. The reason is that only then can he conveniently refer to the
coordinates that make up the corners of the rectangle. If Hagen draws the rectangle first, then he needs to
know the exact size of the Petri net – which he does not.
The solution is to use layers. When the background library is loaded, Hagen can put parts of his picture
inside a {pgfonlayer} environment. Then this part of the picture becomes part of the layer that is given
as an argument to this environment. When the {tikzpicture} environment ends, the layers are put on top
of each other, starting with the background layer. This causes everything drawn on the background layer to
be behind the main text.
2
% Styles as before
\begin{tikzpicture}[bend angle=45]
\node[place] (waiting) {};
\node[place] (critical) [below of=waiting] {};
\node[place] (semaphore) [below of=critical] {};
\node[transition] (leave critical) [right of=critical] {}
edge [pre] (critical)
edge [post,bend right] node[auto,swap] {2} (waiting)
edge [pre, bend left] (semaphore);
\node[transition] (enter critical) [left of=critical] {}
edge [post] (critical)
edge [pre, bend left] (waiting)
edge [post,bend right] (semaphore);
\begin{pgfonlayer}{background}
\filldraw [fill=black!30,draw=red]
(semaphore.south -| enter critical.west)
rectangle (waiting.north -| leave critical.east);
\end{pgfonlayer}
\end{tikzpicture}
3.14 The Complete Code
Hagen has now finally put everything together. Only then does he learn that there is already a library for
drawing Petri nets! It turns out that this library mainly provides the same definitions as Hagen did. For
43
example, it defines a place style in a similar way as Hagen did. Adjusting the code so that it uses the library
shortens Hagen code a bit, as shown in the following.
First, Hagen needs less style definitions, but he still needs to specify the colors of places and transitions.
\tikzstyle{every place}= [minimum size=6mm,thick,draw=blue!75,fill=blue!20]
\tikzstyle{every transition}=[thick,draw=black!75,fill=black!20]
\tikzstyle{red place}= [place,draw=red!75,fill=red!20]
\tikzstyle{every label}= [red]
\begin{tikzpicture}[node distance=1.3cm,>=stealth’,bend angle=45,auto]
Now comes the code for the nets:
s ≤ 3
2
2
\node [place,tokens=1] (w1) {};
\node [place] (c1) [below of=w1] {};
\node [place] (s) [below of=c1,label=above:$s\le 3$] {};
\node [place] (c2) [below of=s] {};
\node [place,tokens=1] (w2) [below of=c2] {};
\node [transition] (e1) [left of=c1] {}
edge [pre,bend left] (w1)
edge [post,bend right] (s)
edge [post] (c1);
\node [transition] (e2) [left of=c2] {}
edge [pre,bend right] (w2)
edge [post,bend left] (s)
edge [post] (c2);
\node [transition] (l1) [right of=c1] {}
edge [pre] (c1)
edge [pre,bend left] (s)
edge [post,bend right] node[swap] {2} (w1);
\node [transition] (l2) [right of=c2] {}
edge [pre] (c2)
edge [pre,bend right] (s)
edge [post,bend left] node {2} (w2);
s ¯ s
2
2
\begin{scope}[xshift=6cm]
\node [place,tokens=1] (w1’) {};
\node [place] (c1’) [below of=w1’] {};
\node [red place] (s1’) [below of=c1’,xshift=-5mm]
[label=left:$s$] {};
\node [red place,tokens=3] (s2’) [below of=c1’,xshift=5mm]
[label=right:$\bar s$] {};
\node [place] (c2’) [below of=s1’,xshift=5mm] {};
\node [place,tokens=1] (w2’) [below of=c2’] {};
\node [transition] (e1’) [left of=c1’] {}
edge [pre,bend left] (w1’)
edge [post] (s1’)
edge [pre] (s2’)
edge [post] (c1’);
\node [transition] (e2’) [left of=c2’] {}
edge [pre,bend right] (w2’)
edge [post] (s1’)
edge [pre] (s2’)
edge [post] (c2’);
\node [transition] (l1’) [right of=c1’] {}
edge [pre] (c1’)
edge [pre] (s1’)
edge [post] (s2’)
edge [post,bend right] node[swap] {2} (w1’);
\node [transition] (l2’) [right of=c2’] {}
edge [pre] (c2’)
edge [pre] (s1’)
edge [post] (s2’)
edge [post,bend left] node {2} (w2’);
\end{scope}
The code for the background and the snake is the following:
44
\draw [-to,thick,snake=snake,segment amplitude=.4mm,segment length=2mm,line after snake=1mm]
([xshift=5mm]s -| l1) -- ([xshift=-5mm]s1’ -| e1’)
node [above=1mm,midway,text width=3cm,text centered]
{replacement of the \textcolor{red}{capacity} by \textcolor{red}{two places}};
\begin{pgfonlayer}{background}
\filldraw [line width=4mm,join=round,black!10]
(w1.north -| l1.east) rectangle (w2.south -| e1.west)
(w1’.north -| l1’.east) rectangle (w2’.south -| e1’.west);
\end{pgfonlayer}
\end{tikzpicture}
45
4 Guidelines on Graphics
The present section is not about pgf or TikZ, but about general guidelines and principles concerning the
creation of graphics for scientific presentations, papers, and books.
The guidelines in this section come from different sources. Many of them are just what I would like to
claim is “common sense,” some reflect my personal experience (though, hopefully, not my personal prefer-
ences), some come from books (the bibliography is still missing, sorry) on graphic design and typography.
The most influential source are the brilliant books by Edward Tufte. While I do not agree with everything
written in these books, many of Tufte’s arguments are so convincing that I decided to repeat them in the
following guidelines.
The first thing you should ask yourself when someone presents a bunch of guidelines is: Should I really
follow these guidelines? This is an important questions, because there are good reasons not to follow general
guidelines.
• The person who setup the guidelines may have had other objectives than you do. For example, a
guideline might say “use the color red for emphasis.” While this guideline makes perfect sense for, say,
a presentation using a projector, red “color” has the opposite effect of “emphasis” when printed using
a black-and-white printer.
Guidelines were almost always setup to address a specific situation. If you are not in this situation,
following a guideline can do more harm than good.
• The basic rule of typography is: “Every rule can be broken, as long as you are aware that you are
breaking a rule.” This rule also applies to graphics. Phrased differently, the basic rule states: “The
only mistakes in typography are things done is ignorance.”
When you are aware of a rule and when you decide that breaking the rule has a desirable effect, break
the rule.
So, before you apply a guideline or choose not to apply it, ask yourself these questions:
1. Does this guideline really address my situation?
2. If you do the opposite a guideline says you should do, will the advantages outweigh the disadvantages
this guideline was supposed to prevent?
4.1 Planning the Time Needed for the Creation of Graphics
When you create a paper with numerous graphics, the time needed to create these graphics becomes an
important factor. How much time should you calculate for the creation of graphics?
As a general rule, assume that a graphic will need as much time to create as would a text of the same
length. For example, when I write a paper, I need about one hour per page for the first draft. Later, I
need between two and four hours per page for revisions. Thus, I expect to need about half an hour for the
creation of a first draft of a half page graphic. Later on, I expect another one to two hours before the final
graphic is finished.
In many publications, even in good journals, the authors and editors have obviously invested a lot of time
on the text, but seem to have spend about five minutes to create all of the graphics. Graphics often seem to
have been added as an “afterthought” or look like a screen shot of whatever the authors’s statistical software
shows them. As will be argued later on, the graphics that programs like gnuplot produce by default are
of poor quality.
Creating informative graphics that help the reader and that fit together with the main text is a difficult,
lengthy process.
• Treat graphics as first-class citizens of your papers. They deserve as much time and energy as the text
does.
• Arguably, the creation of graphics deserves even more time than the writing of the main text since
more attention will be paid to the graphics and they will be looked at first.
• Plan as much time for the creation and revision of a graphic as you would plan for text of the same
size.
• Difficult graphics with a high information density may require even more time.
46
• Very simple graphics will require less time, but most likely you do not want to have “very simple
graphics” in your paper, anyway; just as you would not like to have a “very simple text” of the same
size.
4.2 Workflow for Creating a Graphic
When you write a (scientific) paper, you will most likely follow the following pattern: You have some
results/ideas that you would like to report about. The creation of the paper will typically start with
compiling a rough outline. Then, the different sections are filled with text to create a first draft. This draft
is then revised repeatedly until, often after substantial revision, a final paper results. In a good journal
paper there is typically not be a single sentence that has survived unmodified from the first draft.
Creating a graphics follows the same pattern:
• Decide on what the graphic should communicate. Make this a conscious decision, that is, determine
“What is the graphic supposed to tell the reader?”
• Create an “outline,” that is, the rough overall “shape” of the graphic, containing the most crucial
elements. Often, it is useful to do this using pencil and paper.
• Fill out the finer details of the graphic to create a first draft.
• Revise the graphic repeatedly along with the rest of the paper.
4.3 Linking Graphics With the Main Text
Graphics can be placed at different places in a text. Either, they can be inlined, meaning they are somewhere
“in the middle of the text” or they can be placed in standalone “figures.” Since printers (the people) like to
have their pages “filled,” (both for aesthetic and economic reasons) standalone figures may traditionally be
placed on pages in the document far removed from the main text that refers to them. L
A
T
E
X and T
E
X tend
to encourage this “drifting away” of graphics for technical reasons.
When a graphic is inlined, it will more or less automatically be linked with the main text in the sense
that the labels of the graphic will be implicitly explained by the surrounding text. Also, the main text will
typically make it clear what the graphic is about and what is shown.
Quite differently, a standalone figure will often be viewed at a time when the main text that this graphic
belongs to either has not yet been read or has been read some time ago. For this reason, you should follow
the following guidelines when creating standalone figures:
• Standalone figures should have a caption than should make them “understandable by themselves.”
For example, suppose a graphic shows an example of the different stages of a quicksort algorithm. Then
the figure’s caption should, at the very least, inform the reader that “The figure shows the different
stages of the quicksort algorithm introduced on page xyz.” and not just “Quicksort algorithm.”
• A good caption adds as much context information as possible. For example, you could say: “The
figure shows the different stages of the quicksort algorithm introduced on page xyz. In the first line,
the pivot element 5 is chosen. This causes. . . ” While this information can also be given in the main
text, putting it in the caption will ensure that the context is kept. Do not feel afraid of a 5-line caption.
(Your editor may hate you for this. Consider hating them back.)
• Reference the graphic in your main text as in “For an example of quicksort ‘in action,’ see Figure 2.1
on page xyz.”
• Most books on style and typography recommend that you do not use abbreviations as in “Fig. 2.1”
but write “Figure 2.1.”
The main argument against abbreviations is that “a period is too valuable to waste it on an abbrevi-
ation.” The idea is that a period will make the reader assume that the sentence ends after “Fig” and
it takes a “conscious backtracking” to realize that the sentence did not end after all.
The argument in favor of abbreviations is that they save space.
Personally, I am not really convinced by either argument. On the one hand, I have not yet seen any
hard evidence that abbreviations slow readers down. On the other hand, abbreviating all “Figure” by
“Fig.” is most unlikely to save even a single line in most documents.
I avoid abbreviations.
47
4.4 Consistency Between Graphics and Text
Perhaps the most common “mistake” people do when creating graphics (remember that a “mistake” in
design is always just “ignorance”) is to have a mismatch between the way their graphics look and the way
their text looks.
It is quite common that authors use several different programs for creating the graphics of a paper. An
author might produce some plots using gnuplot, a diagram using xfig, and include an .eps graphic a
coauthor contributed using some unknown program. All these graphics will, most likely, use different line
widths, different fonts, and have different sizes. In addition, authors often use options like [height=5cm]
when including graphics to scale them to some “nice size.”
If the same approach were taken to writing the main text, every section would be written in a different
font at a different size. In some sections all theorems would be underlined, in another they would be printed
all in uppercase letters, and in another in red. In addition, the margins would be different on each page.
Readers and editors would not tolerate a text if it were written in this fashion, but with graphics they
often have to.
To create consistency between graphics and text, stick to the following guidelines:
• Do not scale graphics.
This means that when generating graphics using an external program, create them “at the right size.”
• Use the same font(s) both in graphics and the body text.
• Use the same line width in text and graphics.
The “line width” for normal text is the width of the stem of letters like T. For T
E
X, this is usually
0.4 pt. However, some journals will not accept graphics with a normal line width below 0.5 pt.
• When using colors, use a consistent color coding in the text and in graphics. For example, if red is
supposed to alert the reader to something in the main text, use red also in graphics for important parts
of the graphic. If blue is used for structural elements like headlines and section titles, use blue also for
structural elements of your graphic.
However, graphics may also use a logical intrinsic color coding. For example, no matter what colors
you normally use, readers will generally assume, say, that the color green as “positive, go, ok” and red
as “alert, warning, action.”
Creating consistency when using different graphic programs is almost impossible. For this reason, you
should consider sticking to a single graphic program.
4.5 Labels in Graphics
Almost all graphics will contain labels, that is, pieces of text that explain parts of the graphics. When
placing labels, stick to the following guidelines:
• Follow the rule of consistency when placing labels. You should do so in two ways: First, be consistent
with the main text, that is, use the same font as the main text also for labels. Second, be consistent
between labels, that is, if you format some labels in some particular way, format all labels in this way.
• In addition to using the same fonts in text and graphics, you should also use the same notation. For
example, if you write 1/2 in your main text, also use “1/2” as labels in graphics, not “0.5”. A π is a
“π” and not “3.141”. Finally, e
−iπ
is “e
−iπ
”, not “−1”, let alone “-1”.
• Labels should be legible. They should not only have a reasonably large size, they also should not be
obscured by lines or other text. This also applies to of lines and text behind the labels.
• Labels should be “in place.” Whenever there is enough space, labels should be placed next to the
thing they label. Only if necessary, add a (subdued) line from the label to the labeled object. Try to
avoid labels that only reference explanations in external legends. Reader have to jump back and forth
between the explanation and the object that is described.
• Consider subduing “unimportant” labels using, for example, a gray color. This will keep the focus on
the actual graphic.
48
4.6 Plots and Charts
One of the most frequent kind of graphics, especially in scientific papers, are plots. They come in a large
variety, including simple line plots, parametric plots, three dimensional plots, pie charts, and many more.
Unfortunately, plots are notoriously hard to get right. Partly, the default settings of programs like
gnuplot or Excel are to blame for this since these programs make it very convenient to create bad plots.
The first question you should ask yourself when creating a plot is the following:
• Are there enough data points to merit a plot?
If the answer is “not really,” use a table.
A typical situation where a plot is unnecessary is when people present a few numbers in a bar diagram.
Here is a real-life example: At the end of a seminar a lecturer asked the participants for feedback. Of the 50
participants, 30 returned the feedback form. According to the feedback, three participants considered the
seminar “very good,” nine considered it “good,” ten “ok,” eight “bad,” and no one thought that the seminar
was “very bad.”
A simple way of summing up this information is the following table:
Rating given Participants (out of 50)
who gave this rating
Percentage
“very good” 3 6%
“good” 9 18%
“ok” 10 20%
“bad” 8 16%
“very bad” 0 0%
none 20 40%
What the lecturer did was to visualize the data using a 3D bar diagram. It looked like this:
0
20
40
60
80
100
v
e
r
y
g
o
o
d
g
o
o
d
o
k
b
a
d
v
e
r
y
b
a
d
Both the table and the “plot” have about the same size. If your first thought is “the graphic looks nicer
than the table,” try to answer the following questions based on the information in the table or in the graphic:
1. How many participants where there?
2. How many participants returned the feedback form?
3. What percentage of the participants returned the feedback form?
4. How many participants checked “very good”?
5. What percentage out of all participants checked “very good”?
6. Did more than a quarter of the participants check “bad” or “very bad”?
7. What percentage of the participants that returned the form checked “very good”?
Sadly, the graphic does not allow us to answer a single one of these questions. The table answers all of
them directly, except for the last one. In essence, the information density of the graphic is very nearly zero.
The table has a much higher information density; despite the fact that it uses quite a lot of white space to
present a few numbers.
Here is the list of things that went wrong with the 3D-bar diagram:
49
• The whole graphic is dominated by irritating background lines.
• It is not clear what the numbers at the left mean; presumably percentages, but it might also be the
absolute number of participants.
• The labels at the bottom are rotated, making them hard to read.
(In the real presentation that I saw, the text was rendered at a very low resolution with about 10 by
6 pixels per letter with wrong kerning, making the rotated text almost impossible to read.)
• The third dimension adds complexity to the graphic without adding information.
• The three dimensional setup makes it much harder to gauge the height of the bars correctly. Consider
the “bad” bar. It the number this bar stands for more than 20 or less? While the front of the bar is
below the 20 line, the back of the bar (which counts) is above.
• It is impossible to tell which numbers are represented by the bars. Thus, the bars needlessly hide the
information these bars are all about.
• What do the bar heights add up to? Is it 100% or 60%?
• Does the bar for “very bad” represent 0 or 1?
• Why are the bars blue?
You might argue that in the example the exact numbers are not important for the graphic. The important
things is the “message,” which is that there are more “very good” and “good” ratings than “bad” and “very
bad.” However, to convey this message either use a sentence that says so or use a graphic that conveys this
message more clearly:
Ratings given by
50 participants
“ok”: 10 (20%)
none: 20 (40%)
“very good”: 3 (6%)
“good”: 9 (18%)
“bad”: 8 (16%)
“very bad”: 0 (0%)
The above graphic has about the same information density as the table (about the same size and the
same numbers are shown). In addition, one can directly “see” that there are more good or very good ratings
than bad ones. One can also “see” that the number of people who gave no rating at all is not negligible,
which is quite common for feedback forms.
Charts are not always a good idea. Let us look at an example that I redrew from a pie chart in Die Zeit,
June 4th, 2005:
50
Kohle ist am wichtigsten
Energiemix bei der deutschen Stromerzeugung 2004
Gesamte Netto-Stromerzeugung in Prozent, in Milliarden Kilowattstunden (Mrd. kWh)
9,4%
27,8%
25,6% 22,3%
10,4%
Regenerative (53,7 kWh)/davon Wind 4,4% (25,0 kWh)
Kernenergie
(158,4 kWh)
Braunkohle (146,0 kWh) Steinkohle (127,1 kWh)
Erdgas (59,2 kWh)
Mineral¨ olprodukte (9,2 kWh) 1,6%
Sonstige (16,5 kWh) 2,9%
This graphic has been redrawn in TikZ, but the original looks very similar.
At first sight, the graphic looks “nice and informative,” but there are a lot of things that went wrong:
• The chart is three dimensional. However, the shadings add nothing “information-wise,” at best, they
distract.
• In a 3D-pie-chart the relative sizes are very strongly distorted. For example, the area taken up by the
gray color of “Braunkohle” is larger than the area taken up by the green color of “Kernenergie” despite
the fact that the percentage of Braunkohle is less than the percentage of Kernenergie.
• The 3D-distortion gets worse for small areas. The area of “Regenerative” somewhat larger than the
area of “Erdgas.” The area of “Wind” is slightly smaller than the area of “Mineral¨olprodukte” although
the percentage of Wind is nearly three times larger than the percentage of Mineral¨olprodukte.
In the last case, the different sizes are only partly due to distortion. The designer(s) of the original
graphic have also made the “Wind” slice too small, even taking distortion into account. (Just compare
the size of “Wind” to “Regenerative” in general.)
• According to its caption, this chart is supposed to inform us that coal was the most important energy
source in Germany in 2004. Ignoring the strong distortions caused by the superfluous and misleading
3D-setup, it takes quite a while for this message to get across.
Coal as an energy source is split up into two slices: one for “Steinkohle” and one for “Braunkohle”
(two different kinds of coal). When you add them up, you see that the whole lower half of the pie chart
is taken up by coal.
The two areas for the different kinds of coal are not visually linked at all. Rather, two different colors
are used, the labels are on different sides of the graphic. By comparison, “Regenerative” and “Wind”
are very closely linked.
• The color coding of the graphic follows no logical pattern at all. Why is nuclear energy green? Re-
generative energy is light blue, “other sources” are blue. It seems more like a joke that the area for
“Braunkohle” (which literally translates to “brown coal”) is stone gray, while the area for “Steinkohle”
(which literally translates to “stone coal”) is brown.
• The area with the lightest color is used for “Erdgas.” This area stands out most because of the brighter
color. However, for this chart “Erdgas” is not really important at all.
Edward Tufte calls graphics like the above “chart junk.”
Here are a few recommendations that may help you avoid producing chart junk:
• Do not use 3D pie charts. They are evil.
• Consider using a table instead of a pie chart.
• Due not apply colors randomly; use them to direct the readers’s focus and to group things.
• Do not use background patterns, like a crosshatch or diagonal lines, instead of colors. They distract.
Background patterns in information graphics are evil.
51
4.7 Attention and Distraction
Pick up your favorite fiction novel and have a look at a typical page. You will notice that the page is very
uniform. Nothing is there to distract the reader while reading; no large headlines, no bold text, no large
white areas. Indeed, even when the author does wish to emphasize something, this is done using italic
letters. Such letters blend nicely with the main text—at a distance you will not be able to tell whether a
page contains italic letters, but you would notice a single bold word immediately. The reason novels are
typeset this way is the following paradigm: Avoid distractions.
Good typography (like good organization) is something you do not notice. The job of typography is to
make reading the text, that is, “absorbing” its information content, as effortless as possible. For a novel,
readers absorb the content by reading the text line-by-line, as if they were listening to someone telling the
story. In this situation anything on the page that distracts the eye from going quickly and evenly from line
to line will make the text harder to read.
Now, pick up your favorite weekly magazine or newspaper and have a look at a typical page. You will
notice that there is quite a lot “going on” on the page. Fonts are used at different sizes and in different
arrangements, the text is organized in narrow columns, typically interleaved with pictures. The reason
magazines are typeset in this way is another paradigm: Steer attention.
Readers will not read a magazine like a novel. Instead of reading a magazine line-by-line, we use headlines
and short abstracts to check whether we want to read a certain article or not. The job of typography is to
steer our attention to these abstracts and headlines, first. Once we have decided that we want to read an
article, however, we no longer tolerate distractions, which is why the main text of articles is typeset exactly
the same way as a novel.
The two principles “avoid distractions” and “steer attention” also apply to graphics. When you design a
graphic, you should eliminate everything that will “distract the eye.” At the same time, you should try to
actively help the reader “through the graphic” by using fonts/colors/line widths to highlight different parts.
Here is a non-exhaustive list of things that can distract readers:
• Strong contrasts will always be registered first by the eye. For example, consider the following two
grids:
Even though the left grid comes first in our normal reading order, the right one is much more likely to
be seen first: The white-to-black contrast is higher than the gray-to-white contrast. In addition, there
are more “places” adding to the overall contrast in the right grid.
Things like grids and, more generally, help lines usually should not grab the attention of the readers
and, hence, should be typeset with a low contrast to the background. Also, a loosely-spaced grid is
less distracting than a very closely-spaced grid.
• Dashed lines create many points at which there is black-to-white contrast. Dashed or dotted lines can
be very distracting and, hence, should be avoided in general.
Do not use different dashing patterns to differentiate curves in plots. You loose data points this way
and the eye is not particularly good at “grouping things according to a dashing pattern.” The eye is
much better at grouping things according to colors.
• Background patterns filling an area using diagonal lines or horizontal and vertical lines or just dots are
almost always distracting and, usually, serve no real purpose.
• Background images and shadings distract and only seldom add anything of importance to a graphic.
• Cute little cliparts can easily draw attention away from the data.
52
Part II
Installation and Configuration
by Till Tantau
This part explains how the system is installed. Typically, someone has already done so for your system, so
this part can be skipped; but if this is not the case and you are the poor fellow who has to do the installation,
read the present part.
q
a start
q
b
q
d
q
c
q
e
0,1,L
1,1,R
1,1,L
0,1,L
0,1,L
1,0,R
1,1,R
0,1,R
1,0,R
The current candidate for the busy beaver for five states. It
is presumed that this Turing machine writes a maximum
number of 1’s before halting among all Turing machines
with five states and the tape alphabet {0, 1}. Proving this
conjecture is an open research problem.
\begin{tikzpicture}[->,>=stealth’,shorten >=1pt,auto,node distance=2.8cm,semithick]
\tikzstyle{every state}=[fill=red,draw=none,text=white]
\node[initial,state] (A) {$q_a$};
\node[state] (B) [above right of=A] {$q_b$};
\node[state] (D) [below right of=A] {$q_d$};
\node[state] (C) [below right of=B] {$q_c$};
\node[state] (E) [below of=D] {$q_e$};
\path (A) edge node {0,1,L} (B)
edge node {1,1,R} (C)
(B) edge [loop above] node {1,1,L} (B)
edge node {0,1,L} (C)
(C) edge node {0,1,L} (D)
edge [bend left] node {1,0,R} (E)
(D) edge [loop below] node {1,1,R} (D)
edge node {0,1,R} (A)
(E) edge [bend left] node {1,0,R} (A);
\node [right=1cm,text width=8cm,font=\footnotesize] at (C)
{
The current candidate for the busy beaver for five states. It is
presumed that this Turing machine writes a maximum number of
$1$’s before halting among all Turing machines with five states
and the tape alphabet $\{0, 1\}$. Proving this conjecture is an
open research problem.
};
\end{tikzpicture}
53
5 Installation
There are different ways of installing pgf, depending on your system and needs, and you may need to install
other packages as well as, see below. Before installing, you may wish to review the licenses under which the
package is distributed, see Section 6.
Typically, the package will already be installed on your system. Naturally, in this case you do not need
to worry about the installation process at all and you can skip the rest of this section.
5.1 Package and Driver Versions
This documentation is part of version 1.18 of the pgf package. In order to run pgf, you need a reasonably
recent T
E
X installation. When using L
A
T
E
X, you need the following packages installed (newer versions should
also work):
• xcolor version 2.00.
• xkeyval version 1.8, if you wish to use TikZ.
With plain T
E
X, xcolor is not needed, but you obviously do not get its (full) functionality.
Currently, pgf supports the following backend drivers:
• pdftex version 0.14 or higher. Earlier versions do not work.
• dvips version 5.94a or higher. Earlier versions may also work.
For inter-picture connections, you need process pictures using pdftex version 1.40 or higher running
in DVI mode.
• dvipdfm version 0.13.2c or higher. Earlier versions may also work.
For inter-picture connections, you need process pictures using pdftex version 1.40 or higher running
in DVI mode.
• tex4ht version 2003-05-05 or higher. Earlier versions may also work.
• vtex version 8.46a or higher. Earlier versions may also work.
• textures version 2.1 or higher. Earlier versions may also work.
Currently, pgf supports the following formats:
• latex with complete functionality.
• plain with complete functionality, except for graphics inclusion, which works only for pdfT
E
X.
• context with complete functionality, except for graphics inclusion, which works only for pdfT
E
X.
For more details, see Section 7.
5.2 Installing Prebundled Packages
I do not create or manage prebundled packages of pgf, but, fortunately, nice other people do. I cannot give
detailed instructions on how to install these packages, since I do not manage them, but I can tell you were
to find them. If you have a problem with installing, you might wish to have a look at the Debian page or
the MikT
E
X page first.
5.2.1 Debian
The command “aptitude install pgf” should do the trick. Sit back and relax. In detail, the following
packages are installed:
http://packages.debian.org/pgf
http://packages.debian.org/latex-xcolor
54
5.2.2 MiKTeX
For MiKT
E
X, use the update wizard to install the (latest versions of the) packages called pgf, xcolor, and
xkeyval.
5.3 Installation in a texmf Tree
For a permanent installation, you place the files of the the pgf package in an appropriate texmf tree.
When you ask T
E
X to use a certain class or package, it usually looks for the necessary files in so-called
texmf trees. These trees are simply huge directories that contain these files. By default, T
E
X looks for files
in three different texmf trees:
• The root texmf tree, which is usually located at /usr/share/texmf/ or c:\texmf\ or somewhere
similar.
• The local texmf tree, which is usually located at /usr/local/share/texmf/ or c:\localtexmf\ or
somewhere similar.
• Your personal texmf tree, which is usually located in your home directory at ~/texmf/ or
~/Library/texmf/.
You should install the packages either in the local tree or in your personal tree, depending on whether
you have write access to the local tree. Installation in the root tree can cause problems, since an update of
the whole T
E
X installation will replace this whole tree.
5.3.1 Installation that Keeps Everything Together
Once you have located the right texmf tree, you must decide whether you want to install pgf in such a way
that “all its files are kept in one place” or whether you want to be “tds-compliant,” where tds means “T
E
X
directory structure.”
If you want to keep “everything in one place,” inside the texmf tree that you have chosen create a
sub-sub-directory called texmf/tex/generic/pgf or texmf/tex/generic/pgf-1.18, if you prefer. Then
place all files of the pgf package in this directory. Finally, rebuild T
E
X’s filename database. This is done by
running the command texhash or mktexlsr (they are the same). In MikT
E
X, there is a menu option to do
this.
5.3.2 Installation that is TDS-Compliant
While the above installation process is the most “natural” one and although I would like to recommend it
since it makes updating and managing the pgf package easy, it is not tds-compliant. If you want to be
tds-compliant, proceed as follows: (If you do not know what tds-compliant means, you probably do not
want to be tds-compliant.)
The .tar file of the pgf package contains the following files and directories at its root: README, doc,
generic, plain, and latex. You should “merge” each of the four directories with the following directories
texmf/doc, texmf/tex/generic, texmf/tex/plain, and texmf/tex/latex. For example, in the .tar file
the doc directory contains just the directory pgf, and this directory has to be moved to texmf/doc/pgf.
The root README file can be ignored since it is reproduced in doc/pgf/README.
You may also consider keeping everything in one place and using symbolic links to point from the tds-
compliant directories to the central installation.
For a more detailed explanation of the standard installation process of packages, you might wish to
consult http://www.ctan.org/installationadvice/. However, note that the pgf package does not come
with a .ins file (simply skip that part).
5.4 Updating the Installation
To update your installation from a previous version, all you need to do is to replace everything in the directory
texmf/tex/generic/pgf with the files of the new version (or in all the directories where pgf was installed, if
you chose a tds-compliant installation). The easiest way to do this is to first delete the old version and then
proceed as described above. Sometimes, there are changes in the syntax of certain command from version
to version. If things no longer work that used to work, you may wish to have a look at the release notes and
at the change log.
55
6 Licenses and Copyright
6.1 Which License Applies?
Different parts of the pgf package are distributed under different licenses:
1. The code of the package is dual-license. This means that you can decide which license you wish to use
when using the pgf package. The two options are:
(a) You can use the gnu Public License, version 2.
(b) You can use the L
A
T
E
X Project Public License, version 1.3c.
2. The documentation of the package is also dual-license. Again, you can choose between two options:
(a) You can use the gnu Free Documentation License, version 1.2.
(b) You can use the L
A
T
E
X Project Public License, version 1.3c.
The “documentation of the package” refers to all files in the subdirectory doc of the pgf package. A
detailed listing can be found in the file doc/generic/pgf/licenses/manifest-documentation.txt. All
files in other directories are part of the “code of the package.” A detailed listing can be found in the file
doc/generic/pgf/licenses/manifest-code.txt.
In the resest of this section, the licenses are presented. The following text is copyrighted, see the plain
text versions of these licenses in the directory doc/generic/pgf/licenses for details.
The example picture used in this manual, the Brave gnu World logo, is taken from the Brave gnu World
homepage, where it is copyrighted as follows: “Copyright (C) 1999, 2000, 2001, 2002, 2003, 2004 Georg
C. F. Greve. Permission is granted to make and distribute verbatim copies of this transcript as long as the
copyright and this permission notice appear.”
6.2 The GNU Public License, Version 2
6.2.1 Preamble
The licenses for most software are designed to take away your freedom to share and change it. By contrast,
the gnu General Public License is intended to guarantee your freedom to share and change free software—to
make sure the software is free for all its users. This General Public License applies to most of the Free
Software Foundation’s software and to any other program whose authors commit to using it. (Some other
Free Software Foundation software is covered by the gnu Library General Public License instead.) You can
apply it to your programs, too.
When we speak of free software, we are referring to freedom, not price. Our General Public Licenses
are designed to make sure that you have the freedom to distribute copies of free software (and charge for
this service if you wish), that you receive source code or can get it if you want it, that you can change the
software or use pieces of it in new free programs; and that you know you can do these things.
To protect your rights, we need to make restrictions that forbid anyone to deny you these rights or to ask
you to surrender the rights. These restrictions translate to certain responsibilities for you if you distribute
copies of the software, or if you modify it.
For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the
recipients all the rights that you have. You must make sure that they, too, receive or can get the source
code. And you must show them these terms so they know their rights.
We protect your rights with two steps: (1) copyright the software, and (2) offer you this license which
gives you legal permission to copy, distribute and/or modify the software.
Also, for each author’s protection and ours, we want to make certain that everyone understands that
there is no warranty for this free software. If the software is modified by someone else and passed on, we
want its recipients to know that what they have is not the original, so that any problems introduced by
others will not reflect on the original authors’ reputations.
Finally, any free program is threatened constantly by software patents. We wish to avoid the danger
that redistributors of a free program will individually obtain patent licenses, in effect making the program
proprietary. To prevent this, we have made it clear that any patent must be licensed for everyone’s free use
or not licensed at all.
The precise terms and conditions for copying, distribution and modification follow.
56
6.2.2 Terms and Conditions For Copying, Distribution and Modification
0. This License applies to any program or other work which contains a notice placed by the copyright
holder saying it may be distributed under the terms of this General Public License. The “Program”,
below, refers to any such program or work, and a “work based on the Program” means either the
Program or any derivative work under copyright law: that is to say, a work containing the Program
or a portion of it, either verbatim or with modifications and/or translated into another language.
(Hereinafter, translation is included without limitation in the term “modification”.) Each licensee is
addressed as “you”.
Activities other than copying, distribution and modification are not covered by this License; they are
outside its scope. The act of running the Program is not restricted, and the output from the Program
is covered only if its contents constitute a work based on the Program (independent of having been
made by running the Program). Whether that is true depends on what the Program does.
1. You may copy and distribute verbatim copies of the Program’s source code as you receive it, in any
medium, provided that you conspicuously and appropriately publish on each copy an appropriate
copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and
to the absence of any warranty; and give any other recipients of the Program a copy of this License
along with the Program.
You may charge a fee for the physical act of transferring a copy, and you may at your option offer
warranty protection in exchange for a fee.
2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on
the Program, and copy and distribute such modifications or work under the terms of Section 1 above,
provided that you also meet all of these conditions:
(a) You must cause the modified files to carry prominent notices stating that you changed the files
and the date of any change.
(b) You must cause any work that you distribute or publish, that in whole or in part contains or is
derived from the Program or any part thereof, to be licensed as a whole at no charge to all third
parties under the terms of this License.
(c) If the modified program normally reads commands interactively when run, you must cause it,
when started running for such interactive use in the most ordinary way, to print or display an
announcement including an appropriate copyright notice and a notice that there is no warranty
(or else, saying that you provide a warranty) and that users may redistribute the program under
these conditions, and telling the user how to view a copy of this License. (Exception: if the
Program itself is interactive but does not normally print such an announcement, your work based
on the Program is not required to print an announcement.)
These requirements apply to the modified work as a whole. If identifiable sections of that work are
not derived from the Program, and can be reasonably considered independent and separate works in
themselves, then this License, and its terms, do not apply to those sections when you distribute them
as separate works. But when you distribute the same sections as part of a whole which is a work based
on the Program, the distribution of the whole must be on the terms of this License, whose permissions
for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote
it.
Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely
by you; rather, the intent is to exercise the right to control the distribution of derivative or collective
works based on the Program.
In addition, mere aggregation of another work not based on the Program with the Program (or with a
work based on the Program) on a volume of a storage or distribution medium does not bring the other
work under the scope of this License.
3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code
or executable form under the terms of Sections 1 and 2 above provided that you also do one of the
following:
(a) Accompany it with the complete corresponding machine-readable source code, which must be
distributed under the terms of Sections 1 and 2 above on a medium customarily used for software
interchange; or,
57
(b) Accompany it with a written offer, valid for at least three years, to give any third party, for a
charge no more than your cost of physically performing source distribution, a complete machine-
readable copy of the corresponding source code, to be distributed under the terms of Sections 1
and 2 above on a medium customarily used for software interchange; or,
(c) Accompany it with the information you received as to the offer to distribute corresponding source
code. (This alternative is allowed only for noncommercial distribution and only if you received
the program in object code or executable form with such an offer, in accord with Subsubsection
b above.)
The source code for a work means the preferred form of the work for making modifications to it. For
an executable work, complete source code means all the source code for all modules it contains, plus
any associated interface definition files, plus the scripts used to control compilation and installation of
the executable. However, as a special exception, the source code distributed need not include anything
that is normally distributed (in either source or binary form) with the major components (compiler,
kernel, and so on) of the operating system on which the executable runs, unless that component itself
accompanies the executable.
If distribution of executable or object code is made by offering access to copy from a designated place,
then offering equivalent access to copy the source code from the same place counts as distribution of
the source code, even though third parties are not compelled to copy the source along with the object
code.
4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under
this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void,
and will automatically terminate your rights under this License. However, parties who have received
copies, or rights, from you under this License will not have their licenses terminated so long as such
parties remain in full compliance.
5. You are not required to accept this License, since you have not signed it. However, nothing else
grants you permission to modify or distribute the Program or its derivative works. These actions
are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the
Program (or any work based on the Program), you indicate your acceptance of this License to do so,
and all its terms and conditions for copying, distributing or modifying the Program or works based on
it.
6. Each time you redistribute the Program (or any work based on the Program), the recipient automati-
cally receives a license from the original licensor to copy, distribute or modify the Program subject to
these terms and conditions. You may not impose any further restrictions on the recipients’ exercise
of the rights granted herein. You are not responsible for enforcing compliance by third parties to this
License.
7. If, as a consequence of a court judgment or allegation of patent infringement or for any other reason
(not limited to patent issues), conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not excuse you from the conditions
of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this
License and any other pertinent obligations, then as a consequence you may not distribute the Program
at all. For example, if a patent license would not permit royalty-free redistribution of the Program
by all those who receive copies directly or indirectly through you, then the only way you could satisfy
both it and this License would be to refrain entirely from distribution of the Program.
If any portion of this section is held invalid or unenforceable under any particular circumstance, the
balance of the section is intended to apply and the section as a whole is intended to apply in other
circumstances.
It is not the purpose of this section to induce you to infringe any patents or other property right claims
or to contest validity of any such claims; this section has the sole purpose of protecting the integrity of
the free software distribution system, which is implemented by public license practices. Many people
have made generous contributions to the wide range of software distributed through that system in
reliance on consistent application of that system; it is up to the author/donor to decide if he or she is
willing to distribute software through any other system and a licensee cannot impose that choice.
This section is intended to make thoroughly clear what is believed to be a consequence of the rest of
this License.
58
8. If the distribution and/or use of the Program is restricted in certain countries either by patents or by
copyrighted interfaces, the original copyright holder who places the Program under this License may
add an explicit geographical distribution limitation excluding those countries, so that distribution is
permitted only in or among countries not thus excluded. In such case, this License incorporates the
limitation as if written in the body of this License.
9. The Free Software Foundation may publish revised and/or new versions of the General Public License
from time to time. Such new versions will be similar in spirit to the present version, but may differ in
detail to address new problems or concerns.
Each version is given a distinguishing version number. If the Program specifies a version number of
this License which applies to it and “any later version”, you have the option of following the terms and
conditions either of that version or of any later version published by the Free Software Foundation.
If the Program does not specify a version number of this License, you may choose any version ever
published by the Free Software Foundation.
10. If you wish to incorporate parts of the Program into other free programs whose distribution conditions
are different, write to the author to ask for permission. For software which is copyrighted by the Free
Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this.
Our decision will be guided by the two goals of preserving the free status of all derivatives of our free
software and of promoting the sharing and reuse of software generally.
6.2.3 No Warranty
10. Because the program is licensed free of charge, there is no warranty for the program, to the extent
permitted by applicable law. Except when otherwise stated in writing the copyright holders and/or
other parties provide the program “as is” without warranty of any kind, either expressed or implied,
including, but not limited to, the implied warranties of merchantability and fitness for a particular
purpose. The entire risk as to the quality and performance of the program is with you. Should the
program prove defective, you assume the cost of all necessary servicing, repair or correction.
11. In no event unless required by applicable law or agreed to in writing will any copyright holder, or
any other party who may modify and/or redistribute the program as permitted above, be liable to
you for damages, including any general, special, incidental or consequential damages arising out of the
use or inability to use the program (including but not limited to loss of data or data being rendered
inaccurate or losses sustained by you or third parties or a failure of the program to operate with any
other programs), even if such holder or other party has been advised of the possibility of such damages.
6.3 The L
A
T
E
X Project Public License, Version 1.3c 2006-05-20
6.3.1 Preamble
The L
A
T
E
X Project Public License (lppl) is the primary license under which the the L
A
T
E
X kernel and the
base L
A
T
E
X packages are distributed.
You may use this license for any work of which you hold the copyright and which you wish to distribute.
This license may be particularly suitable if your work is T
E
X-related (such as a L
A
T
E
X package), but it is
written in such a way that you can use it even if your work is unrelated to T
E
X.
The section ‘wheter and how to distribute works under this license’, below, gives instructions,
examples, and recommendations for authors who are considering distributing their works under this license.
This license gives conditions under which a work may be distributed and modified, as well as conditions
under which modified versions of that work may be distributed.
We, the L
A
T
E
X3 Project, believe that the conditions below give you the freedom to make and distribute
modified versions of your work that conform with whatever technical specifications you wish while maintain-
ing the availability, integrity, and reliability of that work. If you do not see how to achieve your goal while
meeting these conditions, then read the document ‘cfgguide.tex’ and ‘modguide.tex’ in the base L
A
T
E
X
distribution for suggestions.
6.3.2 Definitions
In this license document the following terms are used:
Work Any work being distributed under this License.
59
Derived Work Any work that under any applicable law is derived from the Work.
Modification Any procedure that produces a Derived Work under any applicable law – for example, the
production of a file containing an original file associated with the Work or a significant portion of such
a file, either verbatim or with modifications and/or translated into another language.
Modify To apply any procedure that produces a Derived Work under any applicable law.
Distribution Making copies of the Work available from one person to another, in whole or in part. Distri-
bution includes (but is not limited to) making any electronic components of the Work accessible by file
transfer protocols such as ftp or http or by shared file systems such as Sun’s Network File System
(nfs).
Compiled Work A version of the Work that has been processed into a form where it is directly usable on
a computer system. This processing may include using installation facilities provided by the Work,
transformations of the Work, copying of components of the Work, or other activities. Note that
modification of any installation facilities provided by the Work constitutes modification of the Work.
Current Maintainer A person or persons nominated as such within the Work. If there is no such explicit
nomination then it is the ‘Copyright Holder’ under any applicable law.
Base Interpreter A program or process that is normally needed for running or interpreting a part or the
whole of the Work.
A Base Interpreter may depend on external components but these are not considered part of the
Base Interpreter provided that each external component clearly identifies itself whenever it is used
interactively. Unless explicitly specified when applying the license to the Work, the only applicable
Base Interpreter is a ‘L
A
T
E
X-Format’ or in the case of files belonging to the ‘L
A
T
E
X-format’ a program
implementing the ‘T
E
X language’.
6.3.3 Conditions on Distribution and Modification
1. Activities other than distribution and/or modification of the Work are not covered by this license; they
are outside its scope. In particular, the act of running the Work is not restricted and no requirements
are made concerning any offers of support for the Work.
2. You may distribute a complete, unmodified copy of the Work as you received it. Distribution of only
part of the Work is considered modification of the Work, and no right to distribute such a Derived
Work may be assumed under the terms of this clause.
3. You may distribute a Compiled Work that has been generated from a complete, unmodified copy of
the Work as distributed under Clause 2 above, as long as that Compiled Work is distributed in such a
way that the recipients may install the Compiled Work on their system exactly as it would have been
installed if they generated a Compiled Work directly from the Work.
4. If you are the Current Maintainer of the Work, you may, without restriction, modify the Work, thus
creating a Derived Work. You may also distribute the Derived Work without restriction, including
Compiled Works generated from the Derived Work. Derived Works distributed in this manner by the
Current Maintainer are considered to be updated versions of the Work.
5. If you are not the Current Maintainer of the Work, you may modify your copy of the Work, thus
creating a Derived Work based on the Work, and compile this Derived Work, thus creating a Compiled
Work based on the Derived Work.
6. If you are not the Current Maintainer of the Work, you may distribute a Derived Work provided the
following conditions are met for every component of the Work unless that component clearly states in
the copyright notice that it is exempt from that condition. Only the Current Maintainer is allowed to
add such statements of exemption to a component of the Work.
(a) If a component of this Derived Work can be a direct replacement for a component of the Work when
that component is used with the Base Interpreter, then, wherever this component of the Work
identifies itself to the user when used interactively with that Base Interpreter, the replacement
component of this Derived Work clearly and unambiguously identifies itself as a modified version
of this component to the user when used interactively with that Base Interpreter.
60
(b) Every component of the Derived Work contains prominent notices detailing the nature of the
changes to that component, or a prominent reference to another file that is distributed as part of
the Derived Work and that contains a complete and accurate log of the changes.
(c) No information in the Derived Work implies that any persons, including (but not limited to) the
authors of the original version of the Work, provide any support, including (but not limited to)
the reporting and handling of errors, to recipients of the Derived Work unless those persons have
stated explicitly that they do provide such support for the Derived Work.
(d) You distribute at least one of the following with the Derived Work:
i. A complete, unmodified copy of the Work; if your distribution of a modified component
is made by offering access to copy the modified component from a designated place, then
offering equivalent access to copy the Work from the same or some similar place meets this
condition, even though third parties are not compelled to copy the Work along with the
modified component;
ii. Information that is sufficient to obtain a complete, unmodified copy of the Work.
7. If you are not the Current Maintainer of the Work, you may distribute a Compiled Work generated
from a Derived Work, as long as the Derived Work is distributed to all recipients of the Compiled
Work, and as long as the conditions of Clause 6, above, are met with regard to the Derived Work.
8. The conditions above are not intended to prohibit, and hence do not apply to, the modification, by
any method, of any component so that it becomes identical to an updated version of that component
of the Work as it is distributed by the Current Maintainer under Clause 4, above.
9. Distribution of the Work or any Derived Work in an alternative format, where the Work or that Derived
Work (in whole or in part) is then produced by applying some process to that format, does not relax
or nullify any sections of this license as they pertain to the results of applying that process.
10. (a) A Derived Work may be distributed under a different license provided that license itself honors
the conditions listed in Clause 6 above, in regard to the Work, though it does not have to honor
the rest of the conditions in this license.
(b) If a Derived Work is distributed under a different license, that Derived Work must provide suf-
ficient documentation as part of itself to allow each recipient of that Derived Work to honor the
restrictions in Clause 6 above, concerning changes from the Work.
11. This license places no restrictions on works that are unrelated to the Work, nor does this license place
any restrictions on aggregating such works with the Work by any means.
12. Nothing in this license is intended to, or may be used to, prevent complete compliance by all parties
with all applicable laws.
6.3.4 No Warranty
There is no warranty for the Work. Except when otherwise stated in writing, the Copyright Holder provides
the Work ‘as is’, without warranty of any kind, either expressed or implied, including, but not limited to,
the implied warranties of merchantability and fitness for a particular purpose. The entire risk as to the
quality and performance of the Work is with you. Should the Work prove defective, you assume the cost of
all necessary servicing, repair, or correction.
In no event unless required by applicable law or agreed to in writing will The Copyright Holder, or
any author named in the components of the Work, or any other party who may distribute and/or modify
the Work as permitted above, be liable to you for damages, including any general, special, incidental or
consequential damages arising out of any use of the Work or out of inability to use the Work (including, but
not limited to, loss of data, data being rendered inaccurate, or losses sustained by anyone as a result of any
failure of the Work to operate with any other programs), even if the Copyright Holder or said author or said
other party has been advised of the possibility of such damages.
61
6.3.5 Maintenance of The Work
The Work has the status ‘author-maintained’ if the Copyright Holder explicitly and prominently states near
the primary copyright notice in the Work that the Work can only be maintained by the Copyright Holder
or simply that it is ‘author-maintained’.
The Work has the status ‘maintained’ if there is a Current Maintainer who has indicated in the Work
that they are willing to receive error reports for the Work (for example, by supplying a valid e-mail address).
It is not required for the Current Maintainer to acknowledge or act upon these error reports.
The Work changes from status ‘maintained’ to ‘unmaintained’ if there is no Current Maintainer, or the
person stated to be Current Maintainer of the work cannot be reached through the indicated means of
communication for a period of six months, and there are no other significant signs of active maintenance.
You can become the Current Maintainer of the Work by agreement with any existing Current Maintainer
to take over this role.
If the Work is unmaintained, you can become the Current Maintainer of the Work through the following
steps:
1. Make a reasonable attempt to trace the Current Maintainer (and the Copyright Holder, if the two
differ) through the means of an Internet or similar search.
2. If this search is successful, then enquire whether the Work is still maintained.
(a) If it is being maintained, then ask the Current Maintainer to update their communication data
within one month.
(b) If the search is unsuccessful or no action to resume active maintenance is taken by the Cur-
rent Maintainer, then announce within the pertinent community your intention to take over
maintenance. (If the Work is a L
A
T
E
X work, this could be done, for example, by posting to
comp.text.tex.)
3. (a) If the Current Maintainer is reachable and agrees to pass maintenance of the Work to you, then
this takes effect immediately upon announcement.
(b) If the Current Maintainer is not reachable and the Copyright Holder agrees that maintenance of
the Work be passed to you, then this takes effect immediately upon announcement.
4. If you make an ‘intention announcement’ as described in 2b above and after three months your intention
is challenged neither by the Current Maintainer nor by the Copyright Holder nor by other people, then
you may arrange for the Work to be changed so as to name you as the (new) Current Maintainer.
5. If the previously unreachable Current Maintainer becomes reachable once more within three months of
a change completed under the terms of 3b or 4, then that Current Maintainer must become or remain
the Current Maintainer upon request provided they then update their communication data within one
month.
A change in the Current Maintainer does not, of itself, alter the fact that the Work is distributed under the
lppl license.
If you become the Current Maintainer of the Work, you should immediately provide, within the Work,
a prominent and unambiguous statement of your status as Current Maintainer. You should also announce
your new status to the same pertinent community as in 2b above.
6.3.6 Whether and How to Distribute Works under This License
This section contains important instructions, examples, and recommendations for authors who are consid-
ering distributing their works under this license. These authors are addressed as ‘you’ in this section.
6.3.7 Choosing This License or Another License
If for any part of your work you want or need to use distribution conditions that differ significantly from
those in this license, then do not refer to this license anywhere in your work but, instead, distribute your
work under a different license. You may use the text of this license as a model for your own license, but your
license should not refer to the lppl or otherwise give the impression that your work is distributed under the
lppl.
The document ‘modguide.tex’ in the base L
A
T
E
X distribution explains the motivation behind the con-
ditions of this license. It explains, for example, why distributing L
A
T
E
X under the gnu General Public
62
License (gpl) was considered inappropriate. Even if your work is unrelated to L
A
T
E
X, the discussion in
‘modguide.tex’ may still be relevant, and authors intending to distribute their works under any license are
encouraged to read it.
6.3.8 A Recommendation on Modification Without Distribution
It is wise never to modify a component of the Work, even for your own personal use, without also meeting the
above conditions for distributing the modified component. While you might intend that such modifications
will never be distributed, often this will happen by accident – you may forget that you have modified that
component; or it may not occur to you when allowing others to access the modified version that you are
thus distributing it and violating the conditions of this license in ways that could have legal implications
and, worse, cause problems for the community. It is therefore usually in your best interest to keep your copy
of the Work identical with the public one. Many works provide ways to control the behavior of that work
without altering any of its licensed components.
6.3.9 How to Use This License
To use this license, place in each of the components of your work both an explicit copyright notice including
your name and the year the work was authored and/or last substantially modified. Include also a statement
that the distribution and/or modification of that component is constrained by the conditions in this license.
Here is an example of such a notice and statement:
%% pig.dtx
%% Copyright 2005 M. Y. Name
%
% This work may be distributed and/or modified under the
% conditions of the LaTeX Project Public License, either version 1.3
% of this license or (at your option) any later version.
% The latest version of this license is in
% http://www.latex-project.org/lppl.txt
% and version 1.3 or later is part of all distributions of LaTeX
% version 2005/12/01 or later.
%
% This work has the LPPL maintenance status ‘maintained’.
%
% The Current Maintainer of this work is M. Y. Name.
%
% This work consists of the files pig.dtx and pig.ins
% and the derived file pig.sty.
Given such a notice and statement in a file, the conditions given in this license document would apply,
with the ‘Work’ referring to the three files ‘pig.dtx’, ‘pig.ins’, and ‘pig.sty’ (the last being generated
from ‘pig.dtx’ using ‘pig.ins’), the ‘Base Interpreter’ referring to any ‘L
A
T
E
X-Format’, and both ‘Copyright
Holder’ and ‘Current Maintainer’ referring to the person ‘M. Y. Name’.
If you do not want the Maintenance section of lppl to apply to your Work, change ‘maintained’ above
into ‘author-maintained’. However, we recommend that you use ‘maintained’ as the Maintenance section
was added in order to ensure that your Work remains useful to the community even when you can no longer
maintain and support it yourself.
6.3.10 Derived Works That Are Not Replacements
Several clauses of the lppl specify means to provide reliability and stability for the user community. They
therefore concern themselves with the case that a Derived Work is intended to be used as a (compatible or
incompatible) replacement of the original Work. If this is not the case (e.g., if a few lines of code are reused
for a completely different task), then clauses 6b and 6d shall not apply.
6.3.11 Important Recommendations
Defining What Constitutes the Work The lppl requires that distributions of the Work contain all
the files of the Work. It is therefore important that you provide a way for the licensee to determine which
63
files constitute the Work. This could, for example, be achieved by explicitly listing all the files of the Work
near the copyright notice of each file or by using a line such as:
% This work consists of all files listed in manifest.txt.
in that place. In the absence of an unequivocal list it might be impossible for the licensee to determine
what is considered by you to comprise the Work and, in such a case, the licensee would be entitled to make
reasonable conjectures as to which files comprise the Work.
6.4 GNU Free Documentation License, Version 1.2, November 2002
6.4.1 Preamble
The purpose of this License is to make a manual, textbook, or other functional and useful document “free”
in the sense of freedom: to assure everyone the effective freedom to copy and redistribute it, with or without
modifying it, either commercially or noncommercially. Secondarily, this License preserves for the author and
publisher a way to get credit for their work, while not being considered responsible for modifications made
by others.
This License is a kind of “copyleft”, which means that derivative works of the document must themselves
be free in the same sense. It complements the GNU General Public License, which is a copyleft license
designed for free software.
We have designed this License in order to use it for manuals for free software, because free software needs
free documentation: a free program should come with manuals providing the same freedoms that the software
does. But this License is not limited to software manuals; it can be used for any textual work, regardless
of subject matter or whether it is published as a printed book. We recommend this License principally for
works whose purpose is instruction or reference.
6.4.2 Applicability and definitions
This License applies to any manual or other work, in any medium, that contains a notice placed by the
copyright holder saying it can be distributed under the terms of this License. Such a notice grants a world-
wide, royalty-free license, unlimited in duration, to use that work under the conditions stated herein. The
“Document”, below, refers to any such manual or work. Any member of the public is a licensee, and is
addressed as “you”. You accept the license if you copy, modify or distribute the work in a way requiring
permission under copyright law.
A “Modified Version” of the Document means any work containing the Document or a portion of it,
either copied verbatim, or with modifications and/or translated into another language.
A “Secondary Section” is a named appendix or a front-matter section of the Document that deals
exclusively with the relationship of the publishers or authors of the Document to the Document’s overall
subject (or to related matters) and contains nothing that could fall directly within that overall subject.
(Thus, if the Document is in part a textbook of mathematics, a Secondary Section may not explain any
mathematics.) The relationship could be a matter of historical connection with the subject or with related
matters, or of legal, commercial, philosophical, ethical or political position regarding them.
The “Invariant Sections” are certain Secondary Sections whose titles are designated, as being those of
Invariant Sections, in the notice that says that the Document is released under this License. If a section does
not fit the above definition of Secondary then it is not allowed to be designated as Invariant. The Document
may contain zero Invariant Sections. If the Document does not identify any Invariant Sections then there
are none.
The “Cover Texts” are certain short passages of text that are listed, as Front-Cover Texts or Back-
Cover Texts, in the notice that says that the Document is released under this License. A Front-Cover Text
may be at most 5 words, and a Back-Cover Text may be at most 25 words.
A “Transparent” copy of the Document means a machine-readable copy, represented in a format whose
specification is available to the general public, that is suitable for revising the document straightforwardly
with generic text editors or (for images composed of pixels) generic paint programs or (for drawings) some
widely available drawing editor, and that is suitable for input to text formatters or for automatic translation
to a variety of formats suitable for input to text formatters. A copy made in an otherwise Transparent
file format whose markup, or absence of markup, has been arranged to thwart or discourage subsequent
modification by readers is not Transparent. An image format is not Transparent if used for any substantial
amount of text. A copy that is not “Transparent” is called “Opaque”.
64
Examples of suitable formats for Transparent copies include plain ASCII without markup, Texinfo input
format, LaTeX input format, SGML or XML using a publicly available DTD, and standard-conforming
simple HTML, PostScript or PDF designed for human modification. Examples of transparent image formats
include PNG, XCF and JPG. Opaque formats include proprietary formats that can be read and edited
only by proprietary word processors, SGML or XML for which the DTD and/or processing tools are not
generally available, and the machine-generated HTML, PostScript or PDF produced by some word processors
for output purposes only.
The “Title Page” means, for a printed book, the title page itself, plus such following pages as are
needed to hold, legibly, the material this License requires to appear in the title page. For works in formats
which do not have any title page as such, “Title Page” means the text near the most prominent appearance
of the work’s title, preceding the beginning of the body of the text.
A section “Entitled XYZ” means a named subunit of the Document whose title either is precisely
XYZ or contains XYZ in parentheses following text that translates XYZ in another language. (Here XYZ
stands for a specific section name mentioned below, such as “Acknowledgements”, “Dedications”,
“Endorsements”, or “History”.) To “Preserve the Title” of such a section when you modify the
Document means that it remains a section “Entitled XYZ” according to this definition.
The Document may include Warranty Disclaimers next to the notice which states that this License applies
to the Document. These Warranty Disclaimers are considered to be included by reference in this License,
but only as regards disclaiming warranties: any other implication that these Warranty Disclaimers may have
is void and has no effect on the meaning of this License.
6.4.3 Verbatim Copying
You may copy and distribute the Document in any medium, either commercially or noncommercially, pro-
vided that this License, the copyright notices, and the license notice saying this License applies to the
Document are reproduced in all copies, and that you add no other conditions whatsoever to those of this Li-
cense. You may not use technical measures to obstruct or control the reading or further copying of the copies
you make or distribute. However, you may accept compensation in exchange for copies. If you distribute a
large enough number of copies you must also follow the conditions in section 3.
You may also lend copies, under the same conditions stated above, and you may publicly display copies.
6.4.4 Copying in Quantity
If you publish printed copies (or copies in media that commonly have printed covers) of the Document,
numbering more than 100, and the Document’s license notice requires Cover Texts, you must enclose the
copies in covers that carry, clearly and legibly, all these Cover Texts: Front-Cover Texts on the front cover,
and Back-Cover Texts on the back cover. Both covers must also clearly and legibly identify you as the
publisher of these copies. The front cover must present the full title with all words of the title equally
prominent and visible. You may add other material on the covers in addition. Copying with changes limited
to the covers, as long as they preserve the title of the Document and satisfy these conditions, can be treated
as verbatim copying in other respects.
If the required texts for either cover are too voluminous to fit legibly, you should put the first ones listed
(as many as fit reasonably) on the actual cover, and continue the rest onto adjacent pages.
If you publish or distribute Opaque copies of the Document numbering more than 100, you must either
include a machine-readable Transparent copy along with each Opaque copy, or state in or with each Opaque
copy a computer-network location from which the general network-using public has access to download using
public-standard network protocols a complete Transparent copy of the Document, free of added material. If
you use the latter option, you must take reasonably prudent steps, when you begin distribution of Opaque
copies in quantity, to ensure that this Transparent copy will remain thus accessible at the stated location
until at least one year after the last time you distribute an Opaque copy (directly or through your agents or
retailers) of that edition to the public.
It is requested, but not required, that you contact the authors of the Document well before redistributing
any large number of copies, to give them a chance to provide you with an updated version of the Document.
6.4.5 Modifications
You may copy and distribute a Modified Version of the Document under the conditions of sections 2 and
3 above, provided that you release the Modified Version under precisely this License, with the Modified
65
Version filling the role of the Document, thus licensing distribution and modification of the Modified Version
to whoever possesses a copy of it. In addition, you must do these things in the Modified Version:
A. Use in the Title Page (and on the covers, if any) a title distinct from that of the Document, and
from those of previous versions (which should, if there were any, be listed in the History section of the
Document). You may use the same title as a previous version if the original publisher of that version
gives permission.
B. List on the Title Page, as authors, one or more persons or entities responsible for authorship of
the modifications in the Modified Version, together with at least five of the principal authors of the
Document (all of its principal authors, if it has fewer than five), unless they release you from this
requirement.
C. State on the Title page the name of the publisher of the Modified Version, as the publisher.
D. Preserve all the copyright notices of the Document.
E. Add an appropriate copyright notice for your modifications adjacent to the other copyright notices.
F. Include, immediately after the copyright notices, a license notice giving the public permission to use
the Modified Version under the terms of this License, in the form shown in the Addendum below.
G. Preserve in that license notice the full lists of Invariant Sections and required Cover Texts given in the
Document’s license notice.
H. Include an unaltered copy of this License.
I. Preserve the section Entitled “History”, Preserve its Title, and add to it an item stating at least the
title, year, new authors, and publisher of the Modified Version as given on the Title Page. If there is no
section Entitled “History” in the Document, create one stating the title, year, authors, and publisher
of the Document as given on its Title Page, then add an item describing the Modified Version as stated
in the previous sentence.
J. Preserve the network location, if any, given in the Document for public access to a Transparent copy
of the Document, and likewise the network locations given in the Document for previous versions it
was based on. These may be placed in the “History” section. You may omit a network location for a
work that was published at least four years before the Document itself, or if the original publisher of
the version it refers to gives permission.
K. For any section Entitled “Acknowledgements” or “Dedications”, Preserve the Title of the section, and
preserve in the section all the substance and tone of each of the contributor acknowledgements and/or
dedications given therein.
L. Preserve all the Invariant Sections of the Document, unaltered in their text and in their titles. Section
numbers or the equivalent are not considered part of the section titles.
M. Delete any section Entitled “Endorsements”. Such a section may not be included in the Modified
Version.
N. Do not retitle any existing section to be Entitled “Endorsements” or to conflict in title with any
Invariant Section.
O. Preserve any Warranty Disclaimers.
If the Modified Version includes new front-matter sections or appendices that qualify as Secondary
Sections and contain no material copied from the Document, you may at your option designate some or all
of these sections as invariant. To do this, add their titles to the list of Invariant Sections in the Modified
Version’s license notice. These titles must be distinct from any other section titles.
You may add a section Entitled “Endorsements”, provided it contains nothing but endorsements of
your Modified Version by various parties–for example, statements of peer review or that the text has been
approved by an organization as the authoritative definition of a standard.
You may add a passage of up to five words as a Front-Cover Text, and a passage of up to 25 words
as a Back-Cover Text, to the end of the list of Cover Texts in the Modified Version. Only one passage of
Front-Cover Text and one of Back-Cover Text may be added by (or through arrangements made by) any
66
one entity. If the Document already includes a cover text for the same cover, previously added by you or by
arrangement made by the same entity you are acting on behalf of, you may not add another; but you may
replace the old one, on explicit permission from the previous publisher that added the old one.
The author(s) and publisher(s) of the Document do not by this License give permission to use their
names for publicity for or to assert or imply endorsement of any Modified Version.
6.4.6 Combining Documents
You may combine the Document with other documents released under this License, under the terms defined
in section 4 above for modified versions, provided that you include in the combination all of the Invariant
Sections of all of the original documents, unmodified, and list them all as Invariant Sections of your combined
work in its license notice, and that you preserve all their Warranty Disclaimers.
The combined work need only contain one copy of this License, and multiple identical Invariant Sections
may be replaced with a single copy. If there are multiple Invariant Sections with the same name but different
contents, make the title of each such section unique by adding at the end of it, in parentheses, the name of
the original author or publisher of that section if known, or else a unique number. Make the same adjustment
to the section titles in the list of Invariant Sections in the license notice of the combined work.
In the combination, you must combine any sections Entitled “History” in the various original documents,
forming one section Entitled “History”; likewise combine any sections Entitled “Acknowledgements”, and
any sections Entitled “Dedications”. You must delete all sections Entitled “Endorsements”.
6.4.7 Collection of Documents
You may make a collection consisting of the Document and other documents released under this License,
and replace the individual copies of this License in the various documents with a single copy that is included
in the collection, provided that you follow the rules of this License for verbatim copying of each of the
documents in all other respects.
You may extract a single document from such a collection, and distribute it individually under this
License, provided you insert a copy of this License into the extracted document, and follow this License in
all other respects regarding verbatim copying of that document.
6.4.8 Aggregating with independent Works
A compilation of the Document or its derivatives with other separate and independent documents or works,
in or on a volume of a storage or distribution medium, is called an “aggregate” if the copyright resulting from
the compilation is not used to limit the legal rights of the compilation’s users beyond what the individual
works permit. When the Document is included in an aggregate, this License does not apply to the other
works in the aggregate which are not themselves derivative works of the Document.
If the Cover Text requirement of section 3 is applicable to these copies of the Document, then if the
Document is less than one half of the entire aggregate, the Document’s Cover Texts may be placed on covers
that bracket the Document within the aggregate, or the electronic equivalent of covers if the Document is
in electronic form. Otherwise they must appear on printed covers that bracket the whole aggregate.
6.4.9 Translation
Translation is considered a kind of modification, so you may distribute translations of the Document under
the terms of section 4. Replacing Invariant Sections with translations requires special permission from
their copyright holders, but you may include translations of some or all Invariant Sections in addition to the
original versions of these Invariant Sections. You may include a translation of this License, and all the license
notices in the Document, and any Warranty Disclaimers, provided that you also include the original English
version of this License and the original versions of those notices and disclaimers. In case of a disagreement
between the translation and the original version of this License or a notice or disclaimer, the original version
will prevail.
If a section in the Document is Entitled “Acknowledgements”, “Dedications”, or “History”, the require-
ment (section 4) to Preserve its Title (section 1) will typically require changing the actual title.
6.4.10 Termination
You may not copy, modify, sublicense, or distribute the Document except as expressly provided for under
this License. Any other attempt to copy, modify, sublicense or distribute the Document is void, and will
67
automatically terminate your rights under this License. However, parties who have received copies, or rights,
from you under this License will not have their licenses terminated so long as such parties remain in full
compliance.
6.4.11 Future Revisions of this License
The Free Software Foundation may publish new, revised versions of the GNU Free Documentation License
from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail
to address new problems or concerns. See http://www.gnu.org/copyleft/.
Each version of the License is given a distinguishing version number. If the Document specifies that
a particular numbered version of this License “or any later version” applies to it, you have the option of
following the terms and conditions either of that specified version or of any later version that has been
published (not as a draft) by the Free Software Foundation. If the Document does not specify a version
number of this License, you may choose any version ever published (not as a draft) by the Free Software
Foundation.
6.4.12 Addendum: How to use this License for your documents
To use this License in a document you have written, include a copy of the License in the document and put
the following copyright and license notices just after the title page:
Copyright c _year your name. Permission is granted to copy, distribute and/or modify this
document under the terms of the GNU Free Documentation License, Version 1.2 or any later
version published by the Free Software Foundation; with no Invariant Sections, no Front-Cover
Texts, and no Back-Cover Texts. A copy of the license is included in the section entitled “GNU
Free Documentation License”.
If you have Invariant Sections, Front-Cover Texts and Back-Cover Texts, replace the “with . . . Texts.”
line with this:
with the Invariant Sections being list their titles, with the Front-Cover Texts being list,
and with the Back-Cover Texts being list.
If you have Invariant Sections without Cover Texts, or some other combination of the three, merge those
two alternatives to suit the situation.
If your document contains nontrivial examples of program code, we recommend releasing these examples
in parallel under your choice of free software license, such as the GNU General Public License, to permit
their use in free software.
68
7 Input and Output Formats
T
E
X was designed to be a flexible system. This is true both for the input for T
E
X as well as for the output.
The present section explains which input formats there are and how they are supported by pgf. It also
explains which different output formats can be produced.
7.1 Supported Input Formats
T
E
X does not prescribe exactly how your input should be formatted. While it is customary that, say,
an opening brace starts a scope in T
E
X, this is by no means necessary. Likewise, it is customary that
environments start with \begin, but T
E
X could not really care less about the exact command name.
Even though T
E
X can be reconfigured, users can not. For this reason, certain input formats specify a set
of commands and conventions how input for T
E
X should be formatted. There are currently three “major”
formats: Donald Knuth’s original plain T
E
X format, Leslie Lamport’s popular L
A
T
E
X format, and Hans
Hangen’s ConT
E
Xt format.
7.1.1 Using the L
A
T
E
X Format
Using pgf and TikZ with the L
A
T
E
X format is easy: You say \usepackage{pgf} or \usepackage{tikz}.
Usually, that is all you need to do, all configuration will be done automatically and (hopefully) correctly.
The style files used for the L
A
T
E
X format reside in the subdirectory latex/pgf/ of the pgf-system.
Mainly, what these files do is to include files in the directory generic/pgf. For example, here is the content
of the file latex/pgf/frontends/tikz.sty:
% Copyright 2006 by Till Tantau
%
% This file may be distributed and/or modified
%
% 1. under the LaTeX Project Public License and/or
% 2. under the GNU Public License.
%
% See the file doc/generic/pgf/licenses/LICENSE for more details.
\RequirePackage{pgf,calc,pgffor,xkeyval}
\input{tikz.code.tex}
\endinput
The files in the generic/pgf directory do the actual work.
7.1.2 Using the Plain T
E
X Format
When using the plain T
E
X format, you say \input{pgf.tex} or \input{tikz.tex}. Then, instead of
\begin{pgfpicture} and \end{pgfpicture} you use \pgfpicture and \endpgfpicture.
Unlike for the L
A
T
E
X format, pgf is not as good at discerning the appropriate configuration for the plain
T
E
X format. In particular, it can only automatically determine the correct output format if you use pdftex
or tex plus dvips. For all other output formats you need to set the macro \pgfsysdriver to the correct
value. See the description of using output formats later on.
pgf was originally written for use with L
A
T
E
X and this shows in a number of places. Nevertheless, the
plain T
E
X support is reasonably good.
Like the L
A
T
E
X style files, the plain T
E
X files like tikz.tex also just include the correct tikz.code.tex
file.
7.1.3 Using the ConT
E
Xt Format
When using the ConT
E
Xt format, you say \usemodule[pgf] or \usemodule[tikz]. As for the plain
T
E
X format you also have to replace the start- and end-of-environment tags as follows: Instead of
\begin{pgfpicture} and \end{pgfpicture} you use \startpgfpicture and \stoppgfpicture; similarly,
instead of \begin{tikzpicture} and \end{tikzpicture} you use must now use \starttikzpicture and
\stoptikzpicture; and so on for other environments.
The ConT
E
Xt support is very similar to the plain T
E
X support, so the same restrictions apply: You may
have to set the output format directly and graphics inclusion may be a problem.
69
In addition to pgf and tikz there also exist modules like pgfcore, pgfbaseimage, pgflibrarysnakes
and so on. To use them, you may need to include the module pgfmod first (the modules pgf and tikz both
include pgfmod for you, so typically you can skip this). This special module is necessary since ConT
E
Xt
satanically restricts the length of module names to 6 characters and pgf’s long names are mapped to cryptic
6-letter-names for you by the module pgfmod.
7.2 Supported Output Formats
An output format is a format in which T
E
X outputs the text it has typeset. Producing the output is
(conceptually) a two-stage process:
1. T
E
X typesets your text and graphics. The result of this typesetting is mainly a long list of letter–
coordinate pairs, plus (possibly) some “special” commands. This long list of pairs is written to some-
thing called a .dvi-file.
2. Some other program reads this .dvi-file and translates the letter–coordinate pairs into, say, PostScript
commands for placing the given letter at the given coordinate.
The classical example of this process is the combination of latex and dvips. The latex program (which
is just the tex program called with the L
A
T
E
X-macros preinstalled) produces a .dvi-file as its output. The
dvips program takes this output and produces a .ps-file (a PostScript) file. Possibly, this file is further
converted using, say, ps2pdf, whose name is supposed to mean “PostScript to PDF.” Another example of
programs using this process is the combination of tex and dvipdfm. The dvipdfm program takes a .dvi-
file as input and translates the letter–coordinate pairs therein into pdf-commands, resulting in a .pdf file
directly. Finally, the tex4ht is also a program that takes a .dvi-file and produces an output, this time it
is a .html file. The programs pdftex and pdflatex are special: They directly produce a .pdf-file without
the intermediate .dvi-stage. However, from the programmer’s point of view they behave exactly as if there
where an intermediate stage.
Normally, T
E
X only produces letter–coordinate pairs as its “output.” This obviously makes is difficult
to draw, say, a curve. For this, “special” commands can be used. Unfortunately, these special commands
are not the same for the different programs that process the .dvi-file. Indeed, every program that takes a
.dvi-file as input has a totally different syntax for the special commands.
One of the main jobs of pgf is to “abstract way” the difference in the syntax of the different programs.
However, this means that support for each program has to be “programmed,” which is a time-consuming
and complicated process.
7.2.1 Selecting the Backend Driver
When T
E
X typesets your document, it does not know which program you are going to use to transform the
.dvi-file. If your .dvi-file does not contain any special commands, this would be fine; but these days almost
all .dvi-files contain lots of special commands. It is thus necessary to tell T
E
X which program you are going
to use later on.
Unfortunately, there is no “standard” way of telling this to T
E
X. For the L
A
T
E
X format a sophisticated
mechanism exists inside the graphics package and pgf plugs into this mechanism. For other formats and
when this plugging does not work as expected, it is necessary to tell pgf directly which program you are
going to use. This is done by redefining the macro \pgfsysdriver to an appropriate value before you load
pgf. If you are going to use the dvips program, you set this macro to the value pgfsys-dvips.def; if
you use pdftex or pdflatex, you set it to pgfsys-pdftex.def; and so on. In the following, details of the
support of the different programs are discussed.
7.2.2 Producing PDF Output
pgf supports three programs that produce pdf output (pdf means “portable document format” and was
invented by the Adobe company): dvipdfm, pdftex, and vtex. The pdflatex program is the same as the
pdftex program: it uses a different input format, but the output is exactly the same.
File pgfsys-pdftex.def
This is the driver file for use with pdfT
E
X, that is, with the pdftex or pdflatex command. It includes
pgfsys-common-pdf.def.
This driver has the “complete” functionality. This means, everything pgf “can do at all” is implemented
in this driver.
70
File pgfsys-dvipdfm.def
This is a driver file for use with (la)tex followed by dvipdfm. It includes pgfsys-common-pdf.def.
This driver supports most of pgf’s features, but there are some restrictions:
1. In L
A
T
E
X mode it uses graphicx for the graphics inclusion and does not support masking.
2. In plain T
E
X mode it does not support image inclusion.
3. For remembering of pictures (inter-picture connections) you need to use a recent version of pdftex
running in DVI-mode.
File pgfsys-vtex.def
This is the driver file for use with the commercial vtex program. Even though it produces pdf output,
it includes pgfsys-common-postscript.def. Note that the vtex program can produce both Postscript
and pdf output, depending on the command line parameters. However, whether you produce Postscript
or pdf output does not change anything with respect to the driver.
This driver supports most of pgf’s features, except for the following restrictions:
1. In L
A
T
E
X mode it uses graphicx for the graphics inclusion and does not support masking.
2. In plain T
E
X mode it does not support image inclusion.
3. Shading is fully implemented, but yields the same quality as the implementation for dvips.
4. Opacity is not supported.
5. Remembering of pictures (inter-picture connections) is not supported.
It is also possible to produce a .pdf-file by first producing a PostScript file (see below) and then using a
PostScript-to-pdf conversion program like ps2pdf or the Acrobat Distiller.
7.2.3 Producing PostScript Output
File pgfsys-dvips.def
This is a driver file for use with (la)tex followed by dvips. It includes pgfsys-common-postscript.def.
This driver also supports most of pgf’s features, except for the following restrictions:
1. In L
A
T
E
X mode it uses graphicx for the graphics inclusion and does not support masking.
2. In plain T
E
X mode it does not support image inclusion.
3. Shading is fully implemented, but the results will not be as good as with a driver producing .pdf
as output.
4. Opacity works only in conjunction with newer versions of GhostScript.
5. For remembering of pictures (inter-picture connections) you need to use a recent version of pdftex
running in DVI-mode.
File pgfsys-textures.def
This is a driver file for use with the textures program. It includes pgfsys-common-postscript.def.
This driver has exactly the same restrictions as the driver for dvips.
You can also use the vtex program together with pgfsys-vtex.def to produce Postscript output.
7.2.4 Producing HTML / SVG Output
The tex4ht program converts .dvi-files to .html-files. While the html-format cannot be used to draw
graphics, the svg-format can. Using the following driver, you can ask pgf to produce an svg-picture for
each pgf graphic in your text.
File pgfsys-tex4ht.def
This is a driver file for use with the tex4ht program. It includes pgfsys-common-svg.def.
When using this driver you should be aware of the following restrictions:
1. In L
A
T
E
X mode it uses graphicx for the graphics inclusion.
71
2. In plain T
E
X mode it does not support image inclusion.
3. Remembering of pictures (inter-picture connections) is not supported.
4. Text inside pgfpictures is not supported very well. The reason is that the svg specification
currently does not support text very well and it is also not possible to correctly “escape back” to
html. All these problems will hopefully disappear in the future, but currently only two kinds of
text work reasonably well: First, plain text without math mode, special characters or anything
else special. Second, very simple mathematical text that contains subscripts or superscripts. Even
then, variables are not correctly set in italics and, in general, text simple does not look very nice.
5. If you use text that contains anything special, even something as simple as $\alpha$, this may
corrupt the graphic since text4ht does not always produce valid xml code. So, once more, stick
to very simple node text inside graphics. Sorry.
6. Unlike for other output formats, the bounding box of a picture “really crops” the picture.
7. Matrices do not work.
The driver basically works as follows: When a {pgfpicture} is started, appropriate \special com-
mands are used to directed the output of tex4ht to a new file called \jobname-xxx.svg, where xxx
is a number that is increased for each graphic. Then, till the end of the picture, each (system layer)
graphic command creates a special that inserts appropriate svg literal text into the output file. The
exact details are a bit complicated since the imaging model and the processing model of PostScript/pdf
and svg are not quite the same; but they are “close enough” for pgf’s purposes.
7.2.5 Producing Perfectly Portable DVI Output
File pgfsys-dvi.def
This is a driver file that can be used with any output driver, except for tex4ht.
The driver will produce perfectly portable .dvi files by composing all pictures entirely of black rectan-
gles, the basic and only graphic shape supported by the T
E
X core. Even straight, but slanted lines are
tricky to get right in this model (they need to be composed of lots of little squares).
Naturally, very little is possible with this driver. In fact, so little is possible that it is easier to list what
is possible:
• Text boxes can be placed in the normal way.
• Lines and curves can be drawn (stroked). If they are not horizontal or vertical, they are composed
of hundred of small rectangles.
• Lines of different width are supported.
• Transformations are supported.
Note that, say, even filling is not supported! (Let alone color or anything fancy.)
This driver has only one real application: It might be useful when you only need horizontal or vertical
lines in a picture. Then, the results are quite satisfactory.
72
Part III
TikZ ist kein Zeichenprogramm
by Till Tantau
α
β
γ
δ
E
F
A B
C D
When we assume that AB and CD
are parallel, i. e., AB | CD, then α = δ
and β = γ.
\begin{tikzpicture}
\draw[fill=yellow] (0,0) -- (60:.75cm) arc (60:180:.75cm);
\draw(120:0.4cm) node {$\alpha$};
\draw[fill=green!30] (0,0) -- (right:.75cm) arc (0:60:.75cm);
\draw(30:0.5cm) node {$\beta$};
\begin{scope}[shift={(60:2cm)}]
\draw[fill=green!30] (0,0) -- (180:.75cm) arc (180:240:.75cm);
\draw (30:-0.5cm) node {$\gamma$};
\draw[fill=yellow] (0,0) -- (240:.75cm) arc (240:360:.75cm);
\draw (-60:0.4cm) node {$\delta$};
\end{scope}
\begin{scope}[thick]
\draw (60:-1cm) node[fill=white] {$E$} -- (60:3cm) node[fill=white] {$F$};
\draw[red] (-2,0) node[left] {$A$} -- (3,0) node[right]{$B$};
\draw[blue,shift={(60:2cm)}] (-3,0) node[left] {$C$} -- (2,0) node[right]{$D$};
\draw[shift={(60:1cm)},xshift=4cm]
node [right,text width=6cm,rounded corners,fill=red!20,inner sep=1ex]
{
When we assume that $\color{red}AB$ and $\color{blue}CD$ are
parallel, i.\,e., ${\color{red}AB} \mathbin{\|} \color{blue}CD$,
then $\alpha = \delta$ and $\beta = \gamma$.
};
\end{scope}
\end{tikzpicture}
73
8 Design Principles
This section describes the design principles behind the TikZ frontend, where TikZ means “TikZ ist kein
Zeichenprogramm.” To use TikZ, as a L
A
T
E
X user say \usepackage{tikz} somewhere in the preamble, as
a plain T
E
X user say \input tikz.tex. TikZ’s job is to make your life easier by providing an easy-to-learn
and easy-to-use syntax for describing graphics.
The commands and syntax of TikZ were influenced by several sources. The basic command names and
the notion of path operations is taken from metafont, the option mechanism comes from pstricks, the
notion of styles is reminiscent of svg. To make it all work together, some compromises were necessary. I
also added some ideas of my own, like meta-arrows and coordinate transformations.
The following basic design principles underlie TikZ:
1. Special syntax for specifying points.
2. Special syntax for path specifications.
3. Actions on paths.
4. Key-value syntax for graphic parameters.
5. Special syntax for nodes.
6. Special syntax for trees.
7. Grouping of graphic parameters.
8. Coordinate transformation system.
8.1 Special Syntax For Specifying Points
TikZ provides a special syntax for specifying points and coordinates. In the simplest case, you provide two
T
E
X dimensions, separated by commas, in round brackets as in (1cm,2pt).
You can also specify a point in polar coordinates by using a colon instead of a comma as in (30:1cm),
which means “1cm in a 30 degrees direction.”
If you do not provide a unit, as in (2,1), you specify a point in pgf’s xy-coordinate system. By default,
the unit x-vector goes 1cm to the right and the unit y-vector goes 1cm upward.
By specifying three numbers as in (1,1,1) you specify a point in pgf’s xyz-coordinate system.
It is also possible to use an anchor of a previously defined shape as in (first node.south).
You can add two plus signs before a coordinate as in ++(1cm,0pt). This means “1cm to the
right of the last point used.” This allows you to easily specify relative movements. For example,
(1,0) ++(1,0) ++(0,1) specifies the three coordinates (1,0), then (2,0), and (2,1).
Finally, instead of two plus signs, you can also add a single one. This also specifies a point in a relative
manner, but it does not “change” the current point used in subsequent relative commands. For example,
(1,0) +(1,0) +(0,1) specifies the three coordinates (1,0), then (2,0), and (1,1).
8.2 Special Syntax For Path Specifications
When creating a picture using TikZ, your main job is the specification of paths. A path is a series of straight
or curved lines, which need not be connected. TikZ makes it easy to specify paths, partly using the syntax
of metapost. For example, to specify a triangular path you use
(5pt,0pt) -- (0pt,0pt) -- (0pt,5pt) -- cycle
and you get when you draw this path.
8.3 Actions on Paths
A path is just a series of straight and curved lines, but it is not yet specified what should happen with it.
One can draw a path, fill a path, shade it, clip it, or do any combination of these. Drawing (also known
as stroking) can be thought of as taking a pen of a certain thickness and moving it along the path, thereby
drawing on the canvas. Filling means that the interior of the path is filled with a uniform color. Obviously,
filling makes sense only for closed paths and a path is automatically closed prior to filling, if necessary.
74
Given a path as in \path (0,0) rectangle (2ex,1ex);, you can draw it by adding the draw option as in
\path[draw] (0,0) rectangle (2ex,1ex);, which yields . The \draw command is just an abbreviation
for \path[draw]. To fill a path, use the fill option or the \fill command, which is an abbreviation for
\path[fill]. The \filldraw command is an abbreviation for \path[fill,draw]. Shading is caused by
the shade option (there are \shade and \shadedraw abbreviations) and clipping by the clip option. There
is is also a \clip command, which does the same as \path[clip], but not commands like \drawclip. Use,
say, \draw[clip] or \path[draw,clip] instead.
All of these commands can only be used inside {tikzpicture} environments.
TikZ allows you to use different colors for filling and stroking.
8.4 Key-Value Syntax for Graphic Parameters
Whenever TikZ draws or fills a path, a large number of graphic parameters influenced the rendering. Ex-
amples include the colors used, the dashing pattern, the clipping area, the line width, and many others. In
TikZ, all these options are specified as lists of so called key-value pairs, as in color=red, that are passed
as optional parameters to the path drawing and filling commands. This usage is similar to pstricks. For
example, the following will draw a thick, red triangle;
\tikz \draw[line width=2pt,color=red] (1,0) -- (0,0) -- (1,0) -- cycle;
8.5 Special Syntax for Specifying Nodes
TikZ introduces a special syntax for adding text or, more generally, nodes to a graphic. When you specify
a path, add nodes as in the following example:
text
\tikz \draw (1,1) node {text} -- (2,2);
Nodes are inserted at the current position of the path, but only after the path has been rendered. When
special options are given, as in \draw (1,1) node[circle,draw] {text};, the text is not just put at the
current position. Rather, it is surrounded by a circle and this circle is “drawn.”
You can add a name to a node for later reference either by using the option name=¸node name) or by
stating the node name in parentheses outside the text as in node[circle](name){text}.
Predefined shapes include rectangle, circle, and ellipse, but it is possible (though a bit challenging)
to define new shapes.
8.6 Special Syntax for Specifying Trees
In addition to the “node syntax,” TikZ also introduces a special syntax for drawing trees. The syntax is
intergrated with the special node syntax and only few new commands need to be remebered. In essence, a
node can be followed by any number of children, each introduced by the keyword child. The children are
nodes themselves, each of which may have children in turn.
root
left right
child child
\begin{tikzpicture}
\node {root}
child {node {left}}
child {node {right}
child {node {child}}
child {node {child}}
};
\end{tikzpicture}
Since trees are made up from nodes, it is possible to use options to modify the way trees are drawn. Here
are two examples of the above tree, redrawn with different options:
75
root
left right
child child
\begin{tikzpicture}[edge from parent fork down]
\tikzstyle{every node}=[fill=red!30,rounded corners]
\tikzstyle{edge from parent}=[red,-o,thick,draw]
\node {root}
child {node {left}}
child {node {right}
child {node {child}}
child {node {child}}
};
\end{tikzpicture}
root
left
right
child
child
\begin{tikzpicture}
[parent anchor=east,child anchor=west,grow=east]
\tikzstyle{every node}=[ball color=red,circle,text=white]
\tikzstyle{edge from parent}=[draw,dashed,thick,red]
\node {root}
child {node {left}}
child {node {right}
child {node {child}}
child {node {child}}
};
\end{tikzpicture}
8.7 Grouping of Graphic Parameters
Graphic parameters should often apply to several path drawing or filling commands. For example, we
may wish to draw numerous lines all with the same line width of 1pt. For this, we put these commands
in a {scope} environment that takes the desired graphic options as an optional parameter. Naturally,
the specified graphic parameters apply only to the drawing and filling commands inside the environment.
Furthermore, nested {scope} environments or individual drawing commands can override the graphic pa-
rameters of outer {scope} environments. In the following example, three red lines, two green lines, and one
blue line are drawn:
\begin{tikzpicture}
\begin{scope}[color=red]
\draw (0mm,10mm) -- (10mm,10mm);
\draw (0mm, 8mm) -- (10mm, 8mm);
\draw (0mm, 6mm) -- (10mm, 6mm);
\end{scope}
\begin{scope}[color=green]
\draw (0mm, 4mm) -- (10mm, 4mm);
\draw (0mm, 2mm) -- (10mm, 2mm);
\draw[color=blue] (0mm, 0mm) -- (10mm, 0mm);
\end{scope}
\end{tikzpicture}
The {tikzpicture} environment itself also behaves like a {scope} environment, that is, you can specify
graphic parameters using an optional argument. These optional apply to all commands in the picture.
8.8 Coordinate Transformation System
TikZ relies entirely on pgf’s coordinate transformation system to perform transformations. pgf also supports
canvas transformations, a more low-level transformation system, but this system is not accessible from TikZ.
There are two reasons for this: First, the canvas transformation must be used with great care and often
results in “bad” graphics with changing line width and text in wrong sizes. Second, pgf looses track of
where nodes and shapes are positioned when canvas transformations are used.
For more details on the difference between coordinate transformations and canvas transformations see
Section 42.4.
76
9 Hierarchical Structures:
Package, Environments, Scopes, and Styles
The present section explains how your files should be structured when you use TikZ. On the top level,
you need to include the tikz package. In the main text, each graphic needs to be put in a {tikzpicture}
environment. Inside these environments, you can use {scope} environments to create internal groups. Inside
the scopes you use \path commands to actually draw something. On all levels (except for the package level),
graphic options can be given that apply to everything within the environment.
9.1 Loading the Package and the Libraries
\usepackage{tikz} % L
A
T
E
X
\input tikz.tex % plain T
E
X
\usemodule[tikz] % ConT
E
Xt
This package does not have any options.
This will automatically load the pgf package and some other stuff that TikZ needs (like the xkeyval
package).
pgf needs to know what T
E
X driver you are intending to use. In most cases pgf is clever enough to
determine the correct driver for you; this is true in particular if you L
A
T
E
X. Currently, the only situation
where pgf cannot know the driver “by itself” is when you use plain T
E
X or ConT
E
Xt together with
dvipdfm. In this case, you have to write \def\pgfsysdriver{pgfsys-dvipdfm.def} before you input
tikz.tex.
\usetikzlibrary{¸list of libraries)}
Once TikZ has been loaded, you can use this command to load further libraries. The list of libraries
should contain the names of libraries separated by commas. Instead of curly braces, you can also use
square brackets, which is something ConT
E
Xt users will like. If you try to load a library a second time,
nothing will happen.
Example: \usetikzlibrary{arrows}
The above command will load a whole bunch of extra arrow tip definitions.
What this command does is to load the file pgflibrarytikz¸library).code.tex for each ¸library) in
the ¸list of libraries). Thus, to write your own library file, all you need to do is to place a file of the
appropriate name somewhere where T
E
X can find it. L
A
T
E
X, plain T
E
X, and ConT
E
Xt users can then
use your library.
9.2 Creating a Picture
9.2.1 Creating a Picture Using an Environment
The “outermost” scope of TikZ is the {tikzpicture} environment. You may give drawing commands only
inside this environment, giving them outside (as is possible in many other packages) will result in chaos.
In TikZ, the way graphics are rendered is strongly influenced by graphic options. For example, there
is an option for setting the color used for drawing, another for setting the color used for filling, and also
more obscure ones like the option for setting the prefix used in the filenames of temporary files written while
plotting functions using an external program. The graphic options are nearly always specified in a so-called
key-value style. (The “nearly always” refers to the name of nodes, which can also be specified differently.)
All graphic options are local to the {tikzpicture} to which they apply.
\begin{tikzpicture}[¸options)]
¸environment contents)
\end{tikzpicture}
All TikZ commands should be given inside this environment, except for the \tikzstyle command.
Unlike other packages, it is not possible to use, say, \pgfpathmoveto outside this environment and
doing so will result in chaos. For TikZ, commands like \path are only defined inside this environment,
so there is little chance that you will do something wrong here.
When this environment is encountered, the ¸options) are parsed. All options given here will apply to
the whole picture.
77
Next, the contents of the environment is processed and the graphic commands therein are put into a
box. Non-graphic text is suppressed as well as possible, but non-pgf commands inside a {tikzpicture}
environment should not produce any “output” since this may totally scramble the positioning system
of the backend drivers. The suppressing of normal text, by the way, is done by temporarily switching
the font to \nullfont. You can, however, “escape back” to normal T
E
X typesetting. This happens, for
example, when you specify a node.
At the end of the environment, pgf tries to make a good guess at a good guess at the bounding box of
the graphic and then resizes the box such that the box has this size. To “make its guess,” everytime pgf
encounters a coordinate, it updates the bound box’s size such that it encompasses all these coordinates.
This will usually give a good approximation at the bounding box, but will not always be accurate. First,
the line thickness is not taken into account. Second, controls points of a curve often lie far “outside” the
curve and make the bounding box too large. In this case, you should use the [use as bounding box]
option.
The following option influences the baseline of the resulting picture:
• baseline=¸dimension or coordinate) Normally, the lower end of the picture is put on the baseline
of the surrounding text. For example, when you give the code \tikz\draw(0,0)circle(.5ex);,
pgf will find out that the lower end of the picture is at −.5ex and that the upper end is at .5ex.
Then, the lower end will be put on the baseline, resulting in the following: .
Using this option, you can specify that the picture should be raised or lowered such that the height
¸dimension) is on the baseline. For example, tikz[baseline=0pt]\draw(0,0)circle(.5ex);
yields since, now, the baseline is on the height of the x-axis. If you omit the ¸dimensions), 0pt
is assumed as default.
This options is often useful for “inlined” graphics as in
A B
$A \mathbin{\tikz[baseline] \draw[->>] (0pt,.5ex) -- (3ex,.5ex);} B$
Instead of a ¸dimension) you can also provide a coordinate in parantheses. Then the effect is to
put the baseline on the y-coordinate that the give ¸coordinate) has at the end of the picture. This
means that, at the end of the picture, the ¸coordinate) is evaluated and then the baseline is set to
the y-coordinate of the resulting point. This makes it easy to reference the y-coordinate of, say,
the base line of nodes.
Hello world.
Hello
\tikz[baseline=(X.base)]
\node [cross out,draw] (X) {world.};
Top align:
Top align:
\tikz[baseline=(current bounding box.north)]
\draw (0,0) rectangle (1cm,1ex);
• execute at begin picture=¸code) This option can be used to install some code that will be
executed at the beginning of the picture. This option must be given in the argument of the
{tikzpicture} environment itself since this option will not have an effect otherwise. After all, the
picture has already “started” later on.
This option is mainly used in styles like the every picture style to execute certain code at the
start of a picture.
• execute at end picture=¸code) This option installs some code that will be executed at the end
of the picture. Using this option multiple times will cause the code to accumulate. This option
must also be given in the optional argument of the {tikzpicture} environment.
X
Y
\begin{tikzpicture}[execute at end picture=%
{
\begin{pgfonlayer}{background}
\path[fill=yellow,rounded corners]
(current bounding box.south west) rectangle
(current bounding box.north east);
\end{pgfonlayer}
}]
\node at (0,0) {X};
\node at (2,1) {Y};
\end{tikzpicture}
78
All options “end” at the end of the picture. To set an option “globally” you can use the following style:
• style=every picture This style is installed at the beginning of each picture.
\tikzstyle{every picture}=[semithick]
In other T
E
X format, you should use instead the following commands:
\tikzpicture[¸options)]
¸environment contents)
\endtikzpicture
This is the plain T
E
X version of the environment.
\starttikzpicture[¸options)]
¸environment contents)
\stoptikzpicture
This is the ConT
E
Xt version of the environment.
9.2.2 Creating a Picture Using a Command
The following two commands are used for “small” graphics.
\tikz[¸options)]{¸commands)}
This command places the ¸commands) inside a {tikzpicture} environment and adds a semicolon at
the end. This is just a convenience.
The ¸commands) may not contain a paragraph (an empty line). This is a precaution to ensure that
users really use this command only for small graphics.
Example: \tikz{\draw (0,0) rectangle (2ex,1ex)} yields
\tikz[¸options)]¸text);
If the ¸text) does not start with an opening brace, the end of the ¸text) is the next semicolon that is
encountered.
Example: \tikz \draw (0,0) rectangle (2ex,1ex); yields
9.2.3 Adding a Background
By default, pictures do not have any background, that is, they are “transparent” on all parts on which you
do not draw anything. You may instead wish to have a colored background behind your picture or a black
frame around it or lines above and below it or some other kind of decoration.
Since backgrounds are often not needed at all, the definition of styles for adding backgrounds has been
put in the library package pgflibrarytikzbackgrounds. This package is documented in Section 20.
9.3 Using Scopes to Structure a Picture
Inside a {tikzpicture} environment you can create scopes using the {scope} environment. This environ-
ment is available only inside the {tikzpicture} environment, so once more, there is little chance of doing
anything wrong.
\begin{scope}[¸options)]
¸environment contents)
\end{scope}
All ¸options) are local to the ¸environment contents). Furthermore, the clipping path is also local to
the environment, that is, any clipping done inside the environment “ends” at its end.
79
\begin{tikzpicture}
\begin{scope}[red]
\draw (0mm,0mm) -- (10mm,0mm);
\draw (0mm,1mm) -- (10mm,1mm);
\end{scope}
\draw (0mm,2mm) -- (10mm,2mm);
\begin{scope}[green]
\draw (0mm,3mm) -- (10mm,3mm);
\draw (0mm,4mm) -- (10mm,4mm);
\draw[blue] (0mm,5mm) -- (10mm,5mm);
\end{scope}
\end{tikzpicture}
The following style influences scopes:
• style=every scope This style is installed at the beginning of every scope. I do not know really
know what this might be good for, but who knows?
The following options are useful for scopes:
• execute at begin scope=¸code) This option install some code that will be executed at the be-
ginning of the scope. This option must be given in the argument of the {scope} environment.
The effect applies only to the current scope, not to subscopes.
• execute at end scope=¸code) This option installs some code that will be executed at the end of
the current scope. Using this option multiple times will cause the code to accumulate. This option
must also be given in the optional argument of the {scope} environment.
Again, the effect applies only to the current scope, not to subscopes.
\scope[¸options)]
¸environment contents)
\endscope
Plain T
E
X version of the environment.
\startscope[¸options)]
¸environment contents)
\stopscope
ConT
E
Xt version of the environment.
9.4 Using Scopes Inside Paths
The \path command, which is described in much more detail in later sections, also takes graphic options.
These options are local to the path. Furthermore, it is possible to create local scopes within a path simply
by using curly braces as in
\tikz \draw (0,0) -- (1,1)
{[rounded corners] -- (2,0) -- (3,1)}
-- (3,0) -- (2,1);
Note that many options apply only to the path as a whole and cannot be scoped in this way. For example,
it is not possible to scope the color of the path. See the explanations in the section on paths for more details.
Finally, certain elements that you specify in the argument to the \path command also take local options.
For example, a node specification takes options. In this case, the options apply only to the node, not to the
surrounding path.
9.5 Using Styles to Manage How Pictures Look
There is a way of organizing sets of graphic options “orthogonally” to the normal scoping mechanism. For
example, you might wish all your “help lines” to be drawn in a certain way like, say, gray and thin (do not
dash them, that distracts). For this, you can use styles.
A style is simply a set of graphic options that is predefined at some point. Once a style has been defined,
it can be used anywhere using the style option:
80
• style=¸style name) invokes all options that are currently set in the ¸style name). An example of a
style is the predefined help lines style, which you should use for lines in the background like grid
lines or construction lines. You can easily define new styles and modify existing ones.
\begin{tikzpicture}
\draw (0,0) grid +(2,2);
\draw[style=help lines] (2,0) grid +(2,2);
\end{tikzpicture}
\tikzstyle¸style name)+=[¸options)]
This command defines the style ¸style name). Whenever it is used using the style=¸style name)
command, the ¸options) will be invoked. It is permissible that a style invokes another style using the
style= command inside the ¸options), which allows you to build hierarchies of styles. Naturally, you
should not create cyclic dependencies.
If the style already has a predefined meaning, it will unceremoniously be redefined without a warning.
\tikzstyle{help lines}=[blue!50,very thin]
\begin{tikzpicture}
\draw (0,0) grid +(2,2);
\draw[style=help lines] (2,0) grid +(2,2);
\end{tikzpicture}
If the optional + is given, the options are added to the existing definition:
\tikzstyle{help lines}+=[dashed]% aaarghhh!!!
\begin{tikzpicture}
\draw (0,0) grid +(2,2);
\draw[style=help lines] (2,0) grid +(2,2);
\end{tikzpicture}
It is also possible to set a style using an option:
• set style={{¸style name)}+=[¸options)]} This option has the same effect as saying \tikzstyle
before the argument of the option.
\begin{tikzpicture}[set style={{help lines}+=[dashed]}]
\draw (0,0) grid +(2,2);
\draw[style=help lines] (2,0) grid +(2,2);
\end{tikzpicture}
81
10 Specifying Coordinates
10.1 Overview
A coordinate is a position on the canvas on which your picture is drawn. TikZ uses a special syntax for specify-
ing coordinates. Coordinates are always put in round brackets. The general syntax is ([¸options)]¸coordinate
specification)).
The ¸coordinate specification) specified coordinates using one of many different possible coordinate sys-
tems. Examples are the Cartesian coordinate system or polar coordinates or spherical coordinates. No
matter which coordinate system is used, in the end, a specific point on the canvas is represented by the
coordinate.
There are two ways of specifying which coordinate system should be used:
Explicitly You can specify the coordinate system explicitly. To do so, you give the name of the coordi-
nate system at the beginning, followed by cs:, which stands for “coordinate system,” followed by a
specification of the coordinate using the key-value syntax. Thus, the general syntax for ¸coordinate
specification) in the explicit case is (¸coordinate system) cs:¸list of key-value pairs specific to the
coordinate system)).
Implicitly The explicit specification is often too verbose when numerous coordinates should be given.
Because of this, for the coordinate systems that you are likely to use often a special syntax is provided.
TikZ will notice when you use a coordinate specified in a special syntax and will choose the correct
coordinate system automatically.
Here is an example in which explicit the coordinate systems are specified explicitly:
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (canvas cs:x=0cm,y=2mm)
-- (canvas polar cs:radius=2cm,angle=30);
\end{tikzpicture}
In the next example, the coordinate systems are implicit:
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0cm,2mm) -- (30:2cm);
\end{tikzpicture}
It is possible to give options that apply only to a single coordinate, although this makes sense for
transformation options only. To give transformation options for a single coordinate, give these options at
the beginning in brackets:
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0,0) -- (1,1);
\draw[red] (0,0) -- ([xshift=3pt] 1,1);
\draw (1,0) -- +(30:2cm);
\draw[red] (1,0) -- +([shift=(135:5pt)] 30:2cm);
\end{tikzpicture}
10.2 Coordinate Systems
10.2.1 Canvas, XYZ, and Polar Coordinate Systems
Let us start with the basic coordinate systems.
Coordinate system canvas
The simplest way of specifying a coordinate is to use the canvas coordinate system. You provide a
dimension d
x
using the x= option and another dimension d
y
using the y= option. The position on the
canvas is located at the position that is d
x
to the right and d
y
above the origin.
82
• x=¸dimension) Distance by which the coordinate is to the right of the origin. You can also write
things like 1cm+2pt since the calc package is used.
• y=¸dimension) Distance by which the coordinate is above the origin.
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\fill (canvas cs:x=1cm,y=1.5cm) circle (2pt);
\fill (canvas cs:x=2cm,y=-5mm+2pt) circle (2pt);
\end{tikzpicture}
To specify a coordinate in the coordinate system implicitly, you use two dimensions that are seperated
by a comma as in (0cm,3pt) or (2cm,\textheight).
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\fill (1cm,1.5cm) circle (2pt);
\fill (2cm,-5mm+2pt) circle (2pt);
\end{tikzpicture}
Coordinate system xyz
The xyz coordinate system allows you to specify a point as a multiple of three vectors called the
x-, y-, and z-vectors. By default, the x-vector points 1cm to the right, the y-vector points 1cm up-
wards, but this can be changed arbitrarily as explained in Section 17.2. The default z-vector points to


1

2
cm, −
1

2
cm

.
To specify the factors by which the vectors should be multiplied before being added, you use the following
three options:
• x=¸factor) Factor by which the x-vector is multiplied. If this option is not given, 0 is used.
• y=¸factor) Works like x.
• z=¸factor) Works like x.
\begin{tikzpicture}[->]
\draw (0,0) -- (xyz cs:x=1);
\draw (0,0) -- (xyz cs:y=1);
\draw (0,0) -- (xyz cs:z=1);
\end{tikzpicture}
This coordinate system can also be selected implicitly. To do so, you just provide two or three comma-
seperated factors (not dimensions).
\begin{tikzpicture}[->]
\draw (0,0) -- (1,0);
\draw (0,0) -- (0,1,0);
\draw (0,0) -- (0,0,1);
\end{tikzpicture}
Coordinate system canvas polar
The canvas polar coordinate system allows you to specify polar coordinates. You provide an angle
using the angle= option and a radius using the radius= option. This yields the point on the canvas
that is at the given radius distance from the origin at the given degree. A degree of zero points to the
right, a degree of 90 upward.
• angle=¸degrees) The angle of the coordinate. The angle must always be given in degrees and
should be between −360 and 720.
• radius=¸dimension) The distance from the origin.
83
• x radius=¸dimension) A polar coordinate is, after all, just a point on a circle of the given ¸radius).
When you provide an x-radius and also a y-radius, you specify an ellipse instead of a circle. The
radius option has the same effect as specifiying identical x radius and y radius options.
• y radius=¸dimension) Works like x radius.
\tikz \draw (0,0) -- (canvas polar cs:angle=30,radius=1cm);
The implicit form for canvas polar coodinates is the following: you specify the angle and the distance,
separated by a colon as in (30:1cm).
\tikz \draw (0cm,0cm) -- (30:1cm) -- (60:1cm) -- (90:1cm)
-- (120:1cm) -- (150:1cm) -- (180:1cm);
Two different radii are specified by writing (30:1cm and 2cm).
For the implicit form, instead of an angle given as a number you can also use certain words. For example,
up is the same as 90, so that you can write \tikz \draw (0,0) -- (2ex,0pt) -- +(up:1ex); and get
. Apart from up you can use down, left, right, north, south, west, east, north east, north west,
south east, south west, all of which have their natural meaning.
Coordinate system xyz polar
This coordinate system work similarly to the canvas polar system. However, the radius and the angle
are interpreted in the xy-coordinate system, not in the canvas system. More detailedly, consider the
circle or ellipse whose half axes are given by the current x-vector and the current y-vector. Then,
consider the point that lies at a given angle on this ellipse, where an angle of zero is the same as the
x-vector and an angle of 90 is the y-vector. Finally, multiply the resulting vector by the given radius
factor. Voil`a.
• angle=¸degrees) The angle of the coordinate interpreted in the ellipse whose axes are the x-vector
and the y-vector.
• radius=¸factor) A factor by which the x-vector and y-vector are multiplied prior to forming the
ellipse.
• x radius=¸dimension) A specific factor by which only the x-vector is multiplied.
• y radius=¸dimension) works like x radius.
\begin{tikzpicture}[x=1.5cm,y=1cm]
\draw[help lines] (0cm,0cm) grid (3cm,2cm);
\draw (0,0) -- (xyz polar cs:angle=0,radius=1);
\draw (0,0) -- (xyz polar cs:angle=30,radius=1);
\draw (0,0) -- (xyz polar cs:angle=60,radius=1);
\draw (0,0) -- (xyz polar cs:angle=90,radius=1);
\draw (xyz polar cs:angle=0,radius=2)
-- (xyz polar cs:angle=30,radius=2)
-- (xyz polar cs:angle=60,radius=2)
-- (xyz polar cs:angle=90,radius=2);
\end{tikzpicture}
The implicit version of this option is the same as the implicit version of canvas polar, only you do not
provide a unit.
\tikz[x={(0cm,1cm)},y={(-1cm,0cm)}]
\draw (0,0) -- (30:1) -- (60:1) -- (90:1)
-- (120:1) -- (150:1) -- (180:1);
Coordinate system xy polar
This is just an alias for xyz polar, which some people might prefer as there is no x-coordinate involved
in the xyz polar coordinates.
84
10.2.2 Barycentric Systems
In the barycentric coordinate system a point is expressed as the linear combination of multiple vectors. The
idea is that you specify vectors v
1
, v
2
, . . . , v
n
and numbers α
1
, α
2
, . . . , α
n
. Then the barycentric coordinate
specified by these vectors and numbers is
α
1
v
1

2
v
2
+ +α
n
v
n
α
1

2
+ +α
n
The barycentric cs allows you to specifiy such coordiantes easily.
Coordinate system barycentric
For this coordinate system, the ¸coordinate specification) should be a comma-separated list of expressions
of the form ¸node name)=¸number). Note that (currently) the list should not contain any spaces before
or after the ¸node name) (unlike normal key-value pairs).
The specified coordinate is now computed as follows: Each pair provides one vector and a number. The
vector is the center anchor of the ¸node name). The number is the ¸number). Note that (currently)
you cannot specify a different anchor, so that in order to use, say, the north anchor of a node you first
have to create a new coordinate at this north anchor. (Using for instance \coordinate(mynorth) at
(mynode.north);.)
content oriented
structure oriented form oriented
PostScript
DVI
PDF
CSS
XML
HTML
T
E
X
L
A
T
E
X
Word
ASCII
\begin{tikzpicture}
\coordinate (content) at (90:3cm);
\coordinate (structure) at (210:3cm);
\coordinate (form) at (-30:3cm);
\node [above] at (content) {content oriented};
\node [below left] at (structure) {structure oriented};
\node [below right] at (form) {form oriented};
\draw [thick,gray] (content.south) -- (structure.north east) -- (form.north west) -- cycle;
\small
\node at (barycentric cs:content=0.5,structure=0.1 ,form=1) {PostScript};
\node at (barycentric cs:content=1 ,structure=0 ,form=0.4) {DVI};
\node at (barycentric cs:content=0.5,structure=0.5 ,form=1) {PDF};
\node at (barycentric cs:content=0 ,structure=0.25,form=1) {CSS};
\node at (barycentric cs:content=0.5,structure=1 ,form=0) {XML};
\node at (barycentric cs:content=0.5,structure=1 ,form=0.4) {HTML};
\node at (barycentric cs:content=1 ,structure=0.2 ,form=0.8) {\TeX};
\node at (barycentric cs:content=1 ,structure=0.6 ,form=0.8) {\LaTeX};
\node at (barycentric cs:content=0.8,structure=0.8 ,form=1) {Word};
\node at (barycentric cs:content=1 ,structure=0.05,form=0.05) {ASCII};
\end{tikzpicture}
10.2.3 Node Coordinate System
In pgf and in TikZ it is quite easy to define a node that you wish to reference at a later point. Once
you have defined a node, there are different ways of referencing points of the node. To do so, you use the
following coordinate system:
85
Coordinate system node
This coordinate system is used to reference a specific point inside or on the border of a previously
defined node. It can be used in different ways, so let us go over them one by one.
You can use three options to specify which coordinate you mean:
• name=¸node name) specifies the node in which you which to specify a coordinate. The ¸node name)
is the name that was previously used to name the node using the name=¸node name) option or the
special node name syntax.
• anchor=¸anchor) specifies an anchor of the node. Here is an example:
class Shape
class Rectangle
class Circle
class Ellipse
\begin{tikzpicture}
\node (shape) at (0,2) [draw] {|class Shape|};
\node (rect) at (-2,0) [draw] {|class Rectangle|};
\node (circle) at (2,0) [draw] {|class Circle|};
\node (ellipse) at (6,0) [draw] {|class Ellipse|};
\draw (node cs:name=circle,anchor=north) |- (0,1);
\draw (node cs:name=ellipse,anchor=north) |- (0,1);
\draw[-open triangle 90] (node cs:name=rect,anchor=north)
|- (0,1) -| (node cs:name=shape,anchor=south);
\end{tikzpicture}
• angle=¸degrees) It is also possible to provide an angle instead of an anchor. This coordinate refers
to a point of the node’s border where a ray shot from the center in the given angle hits the border.
Here is an example:
start
\begin{tikzpicture}
\node (start) [draw,shape=ellipse] {start};
\foreach \angle in {-90, -80, ..., 90}
\draw (node cs:name=start,angle=\angle)
.. controls +(\angle:1cm) and +(-1,0) .. (2.5,0);
\end{tikzpicture}
It is possible to provide neither the anchor= option nor the angle= option. In this case, TikZ will
calculate an appropriate border position for you. Here is an example:
A
n
ellipse
A circle
\begin{tikzpicture}
\path (0,0) node(a) [ellipse,rotate=10,draw] {An ellipse}
(3,-1) node(b) [circle,draw] {A circle};
\draw[thick] (node cs:name=a) -- (node cs:name=b);
\end{tikzpicture}
TikZ will be reasonably clever at determining the border points that you “mean,” but, naturally, this
may fail in some situations. If TikZ fails to determine an appropriate border point, the center will be
used instead.
Automatic computation of anchors works only with the line-to operations --, the vertical/horizontal
versions |- and -|, and with the curve-to operation ... For other path commands, such as parabola or
plot, the center will be used. If this is not desired, you should give a named anchor or an angle anchor.
86
Note that if you use an automatic coordinate for both the start and the end of a line-to, as in
--(node cs:name=b)--, then two border coordinates are computed with a move-to between them.
This is usually exactly what you want.
If you use relative coordinates together with automatic anchor coordinates, the relative coordinates are
computed relative to the node’s center, not relative to the border point. Here is an example:
Text
\tikz \draw (0,0) node(x) [draw] {Text}
rectangle (1,1)
(node cs:name=x) -- +(1,1);
Similarly, in the following examples both control points are (1, 1):
X Y
\tikz \draw (0,0) node(x) [draw] {X}
(2,0) node(y) {Y}
(node cs:name=x) .. controls +(1,1) and +(-1,1) ..
(node cs:name=y);
The implicit way of specifying the node coordinate system is to simply use the name of the node in
parentheses as in (a) or to specify a name together with an anchor or an angle separated by a dot as
in (a.north) or (a.10).
Here is a more complete example:
A
n
ellipse
A circle
A
r
e
c
t
a
n
g
le
A
n
o
t
h
e
r
r
e
c
t
a
n
g
l
e
\begin{tikzpicture}[fill=blue!20]
\draw[style=help lines] (-1,-2) grid (6,3);
\path (0,0) node(a) [ellipse,rotate=10,draw,fill] {An ellipse}
(3,-1) node(b) [circle,draw,fill] {A circle}
(2,2) node(c) [rectangle,rotate=20,draw,fill] {A rectangle}
(5,2) node(d) [rectangle,rotate=-30,draw,fill] {Another rectangle};
\draw[thick] (a.south) -- (b) -- (c) -- (d);
\draw[thick,red,->] (a) |- +(1,3) -| (c) |- (b);
\draw[thick,blue,<->] (b) .. controls +(right:2cm) and +(down:1cm) .. (d);
\end{tikzpicture}
10.2.4 Intersection Coordinate Systems
Often you wish to specify a point that is on the intersection of two lines. For this, the following coordinate
system is useful:
Coordinate system intersection
To specify the intersection of two line, you provide two lines using the following two options:
• first line=(¸first coordinate))--(¸second coordinate))
• second line=(¸first coordinate))--(¸second coordinate))
Note that you have to write -- between the coordinate, but this does not mean that anything is added
to the path. This is simply a special syntax.
The coordinate specified in this way is the intersection of the two lines. If the lines do not meet or if
they are identical and arithmetical overflow error will result.
87
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\draw (0,0) coordinate (A) -- (3,2) coordinate (B)
(1,2) -- (3,0);
\fill[red] (intersection cs:
first line={(A)--(B)},
second line={(1,2)--(3,0)}) circle (2pt);
\end{tikzpicture}
The implicit way of specifying this coordinate system is to write (intersection of ¸p
1
)--¸p
2
) and ¸q
1
)--¸q
2
)). Note that there are no parentheses around the p
i
and q
i
. Thus, you would write
(intersection of A--B and 1,2--3,0).
A frequent special case of intersections is the intersection of a vertical line going through a point p and
a horizontal line going through some other point q. For this situation there is another coordinate system.
Coordinate system perpendicular
This coordinate system works the same way as intersection, only the lines are specified differently:
• horizontal line through=(¸coordinate)) Specifies that one line is a horizontal line that goes
through the given coordinate.
• vertical line through=(¸coordinate)) Specifies that the other line is vertical and goes through
the given coordinate.
The implicit syntax is to write (¸p) |- ¸q)) or (¸q) -| ¸p)).
For example, (2,1 |- 3,4) and (3,4 -| 2,1) both yield the same as (2,4) (provided the xy-
coordinate system has not been modified).
The most useful application of the syntax is to draw a line up to some point on a vertical or horizontal
line. Here is an example:
p
1
p
2
q
1
q
2
\begin{tikzpicture}
\path (30:1cm) node(p1) {$p_1$} (75:1cm) node(p2) {$p_2$};
\draw (-0.2,0) -- (1.2,0) node(xline)[right] {$q_1$};
\draw (2,-0.2) -- (2,1.2) node(yline)[above] {$q_2$};
\draw[->] (p1) -- (p1 |- xline);
\draw[->] (p2) -- (p2 |- xline);
\draw[->] (p1) -- (p1 -| yline);
\draw[->] (p2) -- (p2 -| yline);
\end{tikzpicture}
10.2.5 Defining New Coordinate Systems
While the set of coordinate systems that TikZ can parse via their special syntax is fixed, it is possible and
quite easy to define new explicitly named coordinate systems. For this, the following commands are used:
\tikzdeclarecoordinatesystem{¸name)}{¸code)}
This command declares a new coordinate system named ¸name) that can later on be used by writing
(¸name) cs:¸arguments)). When TikZ encounters a coordinate specified in this way, the ¸arguments)
are passed to ¸code) as argument #1.
It is now the job of ¸code) to make sense of the ¸arguments). At the end of ¸code), the two T
E
X
dimensions \pgf@x and \pgf@y should be have the x- and y-canvas coordinate of the coordinate.
It is not necessary, but customary, to parse ¸arguments) using the key-value syntax. However, you can
also parse it in any way you like.
In the following example, a coordinate system cylindrical is defined.
88
\makeatletter
\define@key{cylindricalkeys}{angle}{\def\myangle{#1}}
\define@key{cylindricalkeys}{radius}{\def\myradius{#1}}
\define@key{cylindricalkeys}{z}{\def\myz{#1}}
\tikzdeclarecoordinatesystem{cylindrical}%
{%
\setkeys{cylindricalkeys}{#1}%
\pgfpointadd{\pgfpointxyz{0}{0}{\myz}}{\pgfpointpolarxy{\myangle}{\myradius}}
}
\begin{tikzpicture}[z=0.2pt]
\draw [->] (0,0,0) -- (0,0,350);
\foreach \num in {0,10,...,350}
\fill (cylindrical cs:angle=\num,radius=1,z=\num) circle (1pt);
\end{tikzpicture}
\tikzaliascoordinatesystem{¸new name)}{¸old name)}
Creates an alias of ¸old name).
10.3 Relative and Incremental Coordinates
You can prefix coordinates by ++ to make them “relative.” A coordinate such as ++(1cm,0pt) means “1cm
to the right of the previous position.” Relative coordinates are often useful in “local” contexts:
\begin{tikzpicture}
\draw (0,0) -- ++(1,0) -- ++(0,1) -- ++(-1,0) -- cycle;
\draw (2,0) -- ++(1,0) -- ++(0,1) -- ++(-1,0) -- cycle;
\draw (1.5,1.5) -- ++(1,0) -- ++(0,1) -- ++(-1,0) -- cycle;
\end{tikzpicture}
Instead of ++ you can also use a single +. This also specifies a relative coordinate, but it does not “update”
the current point for subsequent usages of relative coordinates. Thus, you can use this notation to specify
numerous points, all relative to the same “initial” point:
\begin{tikzpicture}
\draw (0,0) -- +(1,0) -- +(1,1) -- +(0,1) -- cycle;
\draw (2,0) -- +(1,0) -- +(1,1) -- +(0,1) -- cycle;
\draw (1.5,1.5) -- +(1,0) -- +(1,1) -- +(0,1) -- cycle;
\end{tikzpicture}
There is one special situation, where relative coordinates are interpreted differently. If you use a relative
coordinate as a control point of a B´ezier curve, the following rule applies: First, a relative first control point
is taken relative to the beginning of the curve. Second, a relative second control point is taken relative to
the end of the curve. Third, a relative end point of a curve is taken relative to the start of the curve.
This special behavior makes it easy to specify that a curve should “leave or arrives from a certain
direction” at the start or end. In the following example, the curve “leaves” at 30

and “arrives” at 60

:
\begin{tikzpicture}
\draw (1,0) .. controls +(30:1cm) and +(60:1cm) .. (3,-1);
\draw[gray,->] (1,0) -- +(30:1cm);
\draw[gray,<-] (3,-1) -- +(60:1cm);
\end{tikzpicture}
89
11 Syntax for Path Specifications
A path is a series of straight and curved line segments. It is specified following a \path command and the
specification must follow a special syntax, which is described in the subsections of the present section.
\path¸specification);
This command is available only inside a {tikzpicture} environment.
The ¸specification) is a long stream of path operations. Most of these path operations tell TikZ how the
path is build. For example, when you write --(0,0), you use a line-to operation and it means “continue
the path from wherever you are to the origin.”
At any point where TikZ expects a path operation, you can also give some graphic options, which is a
list of options in brackets, such as [rounded corners]. These options can have different effects:
1. Some options take “immediate” effect and apply to all subsequent path operations on the path. For
example, the rounded corners option will round all following corners, but not the corners “before”
and if the sharp corners is given later on the path (in a new set of brackets), the rounding effect
will end.
\tikz \draw (0,0) -- (1,1)
[rounded corners] -- (2,0) -- (3,1)
[sharp corners] -- (3,0) -- (2,1);
Another example are the transformation options, which also apply only to subsequent coordinates.
2. The options that have immediate effect can be “scoped” by putting part of a path in curly braces.
For example, the above example could also be written as follows:
\tikz \draw (0,0) -- (1,1)
{[rounded corners] -- (2,0) -- (3,1)}
-- (3,0) -- (2,1);
3. Some options only apply to the path as a whole. For example, the color= option for determining
the color used for, say, drawing the path always applies to all parts of the path. If several different
colors are given for different parts of the path, only the last one (on the outermost scope) “wins”:
\tikz \draw (0,0) -- (1,1)
[color=red] -- (2,0) -- (3,1)
[color=blue] -- (3,0) -- (2,1);
Most options are of this type. In the above example, we would have had to “split up” the path
into several \path commands:
\tikz{\draw (0,0) -- (1,1);
\draw [color=red] (2,0) -- (3,1);
\draw [color=blue] (3,0) -- (2,1);}
By default, the \path command does “nothing” with the path, it just “throws it away.” Thus, if you
write \path(0,0)--(1,1);, nothing is drawn in your picture. The only effect is that the area occupied
by the picture is (possibly) enlarged so that the path fits inside the area. To actually “do” something
with the path, an option like draw or fill must be given somewhere on the path. Commands like
\draw do this implicitly.
Finally, it is also possible to give node specifications on a path. Such specifications can come at different
locations, but they are always allowed when a normal path operation could follow. A node specification
starts with node. Basically, the effect is to typeset the node’s text as normal T
E
X text and to place it
at the “current location” on the path. The details are explained in Section 13.
Note, however, that the nodes are not part of the path in any way. Rather, after everything has been
done with the path what is specified by the path options (like filling and drawing the path due to a fill
and a draw option somewhere in the ¸specification)), the nodes are added in a post-processing step.
The following style influences scopes:
90
• style=every path This style is installed at the beginning of every path. This can be useful for
(temporarily) adding, say, the draw option to everything in a scope.
\begin{tikzpicture}[fill=examplefill] % only sets the color
\tikzstyle{every path}=[draw] % all paths are drawn
\fill (0,0) rectangle +(1,1);
\shade (2,0) rectangle +(1,1);
\end{tikzpicture}
11.1 The Move-To Operation
The perhaps simplest operation is the move-to operation, which is specified by just giving a coordinate where
a path operation is expected.
\path . . . ¸coordinate) . . . ;
The move-to operation normally starts a path at a certain point. This does not cause a line segment to
be created, but it specifies the starting point of the next segment. If a path is already under construction,
that is, if several segments have already been created, a move-to operation will start a new part of the
path that is not connected to any of the previous segments.
\begin{tikzpicture}
\draw (0,0) --(2,0) (0,1) --(2,1);
\end{tikzpicture}
In the specification (0,0) --(2,0) (0,1) --(2,1) two move-to operations are specified: (0,0) and
(0,1). The other two operations, namely --(2,0) and --(2,1) are line-to operations, described next.
11.2 The Line-To Operation
11.2.1 Straight Lines
\path . . . --¸coordinate) . . . ;
The line-to operation extends the current path from the current point in a straight line to the given
coordinate. The “current point” is the endpoint of the previous drawing operation or the point specified
by a prior move-to operation.
You use two minus signs followed by a coordinate in round brackets. You can add spaces before and
after the --.
When a line-to operation is used and some path segment has just been constructed, for example by
another line-to operation, the two line segments become joined. This means that if they are drawn, the
point where they meet is “joined” smoothly. To appreciate the difference, consider the following two
examples: In the left example, the path consists of two path segments that are not joined, but that
happen to share a point, while in the right example a smooth join is shown.
\begin{tikzpicture}[line width=10pt]
\draw (0,0) --(1,1) (1,1) --(2,0);
\draw (3,0) -- (4,1) -- (5,0);
\useasboundingbox (0,1.5); % make bounding box higher
\end{tikzpicture}
11.2.2 Horizontal and Vertical Lines
Sometimes you want to connect two points via straight lines that are only horizontal and vertical. For this,
you can use two path construction operations.
91
\path . . . -|¸coordinate) . . . ;
This operation means “first horizontal, then vertical.”
A
B
\begin{tikzpicture}
\draw (0,0) node(a) [draw] {A} (1,1) node(b) [draw] {B};
\draw (a.north) |- (b.west);
\draw[color=red] (a.east) -| (2,1.5) -| (b.north);
\end{tikzpicture}
\path . . . |-¸coordinate) . . . ;
This operations means “first vertical, then horizontal.”
11.2.3 Snaked Lines
The line-to operation can not only be used to append straight lines to the path, but also “snaked” lines
(called thus because they look a little bit like snakes seen from above).
TikZ and pgf use a concept that I termed snakes for appending such “squiggly” lines. A snake specifies
a way of extending a path between two points in a “fancy manner.”
Normally, a snake will just connect the start point to the end point without starting new subpaths. Thus,
a path containing a snaked line can, nevetheless, still be used for filling. However, this is not always the
case. Some snakes consist of numerous unconnected segments. “Lines” consisting of such snakes cannot be
used as the borders of enclosed areas.
Here are some examples of snakes in action:
\begin{tikzpicture}[thick]
\draw (0,3) -- (3,3);
\draw[snake=zigzag] (0,2.5) -- (3,2.5);
\draw[snake=brace] (0,2) -- (3,2);
\draw[snake=triangles] (0,1.5) -- (3,1.5);
\draw[snake=coil,segment length=4pt] (0,1) -- (3,1);
\draw[snake=coil,segment aspect=0] (0,.5) -- (3,.5);
\draw[snake=expanding waves,segment angle=7] (0,0) -- (3,0);
\end{tikzpicture}
\begin{tikzpicture}
\filldraw[fill=red!20,snake=bumps] (0,0) rectangle (3,2);
\end{tikzpicture}
\begin{tikzpicture}
\filldraw[fill=blue!20] (0,3)
[snake=saw] -- (3,3)
[snake=coil,segment aspect=0] -- (2,1)
[snake=bumps] -| (0,3);
\end{tikzpicture}
No special path operation is needed to use a snake. Instead, you use the following option to “switch on”
snaking:
• snake=¸snake name) This option causes the snake ¸snake name) to be used for subsequent line-to
operations. So, whenever you use the -- syntax to specify that a straight line should be added to the
path, a snake to this path will be added instead. Snakes will also be used when you use the -| and
|- syntax and also when you use the rectangle operation. Snakes will not be used when you use the
curve-to operation nor when any other “curved” line is added to the path.
This option has to be given anew for each path. However, you can also leave out the ¸snake name). In
this case, the enclosing scope’s ¸snake name) is used. Thus, you can specify a “standard” snake name
for scope and then just say \draw[snake] every time this snake should actually be used.
92
The ¸snake name) none is special. It can be used to switch off snaking after it has been switched on
on a path.
A bit strangely, no valid ¸snake names) are defined by TikZ by default. Instead, you have to include
the library package pgflibrarysnakes. This package defines numerous snakes, see Section 31 for the
complete list.
Most snakes can be configured. For example, for a snake that looks like a sine curve, you might wish to
change the amplitude or the frequency. There are numerous options that influence these parameters. Not
all options apply to all snakes, see Section 31 once more for details.
• gap before snakes=¸dimension) This option allows you to add a certain “gap” to the snake at its
beginning. The snake will not start at the current point; instead the start point of the snake is move
be ¸dimension) in the direction of the target.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\draw[snake=zigzag] (0,1) -- ++(3,1);
\draw[snake=zigzag,gap before snake=1cm] (0,0) -- ++(3,1);
\end{tikzpicture}
• gap after snake=¸dimension) This option has the same effect as gap before snake, only it affects
the end of the snake, which will “end early.”
• gap around snake=¸dimension) This option sets the gap before and after the gap to ¸dimension).
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\draw[snake=brace] (0,1) -- ++(3,1);
\draw[snake=brace,gap around snake=5mm] (0,0) -- ++(3,1);
\end{tikzpicture}
• line before snake=¸dimension) This option works like gap before snake, only it will connect the
current point with a straight line to the start of the snake.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\draw[snake=zigzag] (0,1) -- ++(3,1);
\draw[snake=zigzag,line before snake=1cm] (0,0) -- ++(3,1);
\end{tikzpicture}
• line after snake=¸dimension) Works line gap after snake, only it adds a straight line.
• line around snake=¸dimension) Works line gap around snake, only it adds straight lines.
• raise snake=¸dimension) This option can be used with all snakes. It will offset the snake by “raising”
it by ¸dimension). A negative ¸dimension) will lower the snake. Raising and lowering is always relative
to the line along which the snake is drawn. Here is an example:
A
B
\begin{tikzpicture}
\node (a) {A};
\node (b) at (2,1) {B};
\draw (a) -- (b);
\draw[snake=brace] (a) -- (b);
\draw[snake=brace,raise snake=5pt,red] (a) -- (b);
\end{tikzpicture}
93
• mirror snake This option causes the snake to be “reflected along the path.” This is best understood
by looking at an example:
A
B
\begin{tikzpicture}
\node (a) {A};
\node (b) at (2,1) {B};
\draw (a) -- (b);
\draw[snake=brace] (a) -- (b);
\draw[snake=brace,mirror snake,red,thick] (a) -- (b);
\end{tikzpicture}
This option can be used with every snake and can be combined with the raise snake option.
• segment amplitude=¸dimension) This option sets the “amplitude” of the snake. For a snake that is a
sine wave this would be the amplitude of this line. For other snakes this value typically describes how
far the snakes “rises above” or “falls below” the path. For some snakes, this value is ignored.
A
B
C
\begin{tikzpicture}
\node (a) {A} node (b) at (2,1) {B} node (c) at (2,-1) {C};
\draw[snake=zigzag] (a) -- (b);
\draw[snake=zigzag,segment amplitude=5pt,red,thick] (a) -- (c);
\end{tikzpicture}
• segment length=¸dimension) This option sets the length of each “segment” of a snake. For a sine
wave this would be the wave length, for other snakes it is the length of each “repetitive part” of the
snake.
A
B
C
\begin{tikzpicture}
\node (a) {A} node (b) at (2,1) {B} node (c) at (2,-1) {C};
\draw[snake=zigzag] (a) -- (b);
\draw[snake=zigzag,segment length=20pt,red,thick] (a) -- (c);
\end{tikzpicture}
A
B
C
\begin{tikzpicture}
\node (a) {A} node (b) at (2,1) {B} node (c) at (2,-1) {C};
\draw[snake=bumps] (a) -- (b);
\draw[snake=bumps,segment length=20pt,red,thick] (a) -- (c);
\end{tikzpicture}
• segment object length=¸dimension) This option sets the length of the objects inside each segment
of a snake. This option is only used for snakes in which each segment contains an object like a triangle
or a star.
A
B
C
\begin{tikzpicture}
\node (a) {A} node (b) at (2,1) {B} node (c) at (2,-1) {C};
\draw[snake=triangles] (a) -- (b);
\draw[snake=triangles,segment object length=8pt,red,thick] (a) -- (c);
\end{tikzpicture}
• segment angle=¸degrees) This option sets an angle that is interpreted in a snake-specific way. For
example, the waves and expanding waves snakes interpret this as (half the) opening angle of the wave.
The border snake uses this value for the angle of the little ticks.
94
A B
\begin{tikzpicture}[segment amplitude=10pt]
\node (a) {A} node (b) at (2,0) {B};
\draw[snake=border] (a) -- (b);
\draw[snake=border,segment angle=20,red,thick] (a) -- (b);
\end{tikzpicture}
A B
A B
\begin{tikzpicture}[segment amplitude=10pt]
\node (a) {A} node (b) at (2,0) {B};
\node (a’) at (0,-1) {A} node (b’) at (2,-1) {B};
\draw[snake=expanding waves] (a) -- (b);
\draw[snake=expanding waves,segment angle=20,red,thick] (a’) -- (b’);
\end{tikzpicture}
• segment aspect=¸ratio) This option sets an aspect ratio that is interpreted in a snake-specific way.
For example, for the coils snake this describes the “direction” from which the coil is viewed.
A
B
C
\begin{tikzpicture}[segment amplitude=5pt,segment length=5pt]
\node (a) {A} node (b) at (2,1) {B} node (c) at (2,-1) {C};
\draw[snake=coil] (a) -- (b);
\draw[snake=coil,segment aspect=0,red,thick] (a) -- (c);
\end{tikzpicture}
It is possible to define new snakes, but this cannot be done inside TikZ. You need to use the command
\pgfdeclaresnake from the basic level directly, see Section 46.
The following styles define combinations of segment settings that may be useful:
• style=snake triangles 45 Installs a snake the consists of little triangles with an opening angle of
45

.
• style=snake triangles 60 Installs a snake the consists of little triangles with an opening angle of
60

.
• style=snake triangles 90 Installs a snake the consists of little triangles with an opening angle of
90

.
11.3 The Curve-To Operation
The curve-to operation allows you to extend a path using a B´ezier curve.
\path . . . ..controls¸c)and¸d)..¸y) . . . ;
This operation extends the current path from the current point, let us call it x, via a curve to a the
current point y. The curve is a cubic B´ezier curve. For such a curve, apart from y, you also specify
two control points c and d. The idea is that the curve starts at x, “heading” in the direction of c.
Mathematically spoken, the tangent of the curve at x goes through c. Similarly, the curve ends at y,
“coming from” the other control point, d. The larger the distance between x and c and between d and y,
the larger the curve will be.
If the “and¸d)” part is not given, d is assumed to be equal to c.
\begin{tikzpicture}
\draw[line width=10pt] (0,0) .. controls (1,1) .. (4,0)
.. controls (5,0) and (5,1) .. (4,1);
\draw[color=gray] (0,0) -- (1,1) -- (4,0) -- (5,0) -- (5,1) -- (4,1);
\end{tikzpicture}
As with the line-to operation, it makes a difference whether two curves are joined because they resulted
from consecutive curve-to or line-to operations, or whether they just happen to have the same ending:
95
\begin{tikzpicture}[line width=10pt]
\draw (0,0) -- (1,1) (1,1) .. controls (1,0) and (2,0) .. (2,0);
\draw (3,0) -- (4,1) .. controls (4,0) and (5,0) .. (5,0);
\useasboundingbox (0,1.5); % make bounding box higher
\end{tikzpicture}
11.4 The Cycle Operation
\path . . . --cycle . . . ;
This operation adds a straight line from the current point to the last point specified by a move-to
operation. Note that this need not be the beginning of the path. Furthermore, a smooth join is created
between the first segment created after the last move-to operation and the straight line appended by
the cycle operation.
Consider the following example. In the left example, two triangles are created using three straight lines,
but they are not joined at the ends. In the second example cycle operations are used.
\begin{tikzpicture}[line width=10pt]
\draw (0,0) -- (1,1) -- (1,0) -- (0,0) (2,0) -- (3,1) -- (3,0) -- (2,0);
\draw (5,0) -- (6,1) -- (6,0) -- cycle (7,0) -- (8,1) -- (8,0) -- cycle;
\useasboundingbox (0,1.5); % make bounding box higher
\end{tikzpicture}
11.5 The Rectangle Operation
A rectangle can obviously be created using four straight lines and a cycle operation. However, since rectangles
are needed so often, a special syntax is available for them.
\path . . . rectangle¸corner) . . . ;
When this operation is used, one corner will be the current point, another corner is given by ¸corner),
which becomes the new current point.
\begin{tikzpicture}
\draw (0,0) rectangle (1,1);
\draw (.5,1) rectangle (2,0.5) (3,0) rectangle (3.5,1.5) -- (2,0);
\end{tikzpicture}
11.6 Rounding Corners
All of the path construction operations mentioned up to now are influenced by the following option:
• rounded corners=¸inset) When this option is in force, all corners (places where a line is continued
either via line-to or a curve-to operation) are replaced by little arcs so that the corner becomes smooth.
\tikz \draw [rounded corners] (0,0) -- (1,1)
-- (2,0) .. controls (3,1) .. (4,0);
The ¸inset) describes how big the corner is. Note that the ¸inset) is not scaled along if you use a
scaling option like scale=2.
96
\begin{tikzpicture}
\draw[color=gray,very thin] (10pt,15pt) circle (10pt);
\draw[rounded corners=10pt] (0,0) -- (0pt,25pt) -- (40pt,25pt);
\end{tikzpicture}
You can switch the rounded corners on and off “in the middle of path” and different corners in the
same path can have different corner radii:
\begin{tikzpicture}
\draw (0,0) [rounded corners=10pt] -- (1,1) -- (2,1)
[sharp corners] -- (2,0)
[rounded corners=5pt] -- cycle;
\end{tikzpicture}
Here is a rectangle with rounded corners:
\tikz \draw[rounded corners=1ex] (0,0) rectangle (20pt,2ex);
You should be aware, that there are several pitfalls when using this option. First, the rounded corner
will only be an arc (part of a circle) if the angle is 90

. In other cases, the rounded corner will still be
round, but “not as nice.”
Second, if there are very short line segments in a path, the “rounding” may cause inadverted effects.
In such case it may be necessary to temporarily switch off the rounding using sharp corners.
• sharp corners This options switches off any rounding on subsequent corners of the path.
11.7 The Circle and Ellipse Operations
A circle can be approximated well using four B´ezier curves. However, it is difficult to do so correctly. For
this reason, a special syntax is available for adding such an approximation of a circle to the current path.
\path . . . circle(¸radius)) . . . ;
The center of the circle is given by the current point. The new current point of the path will remain to
be the center of the circle.
\path . . . ellipse(¸half width) and ¸half height)) . . . ;
Note that you can add spaces after ellipse, but you have to place spaces around and.
\begin{tikzpicture}
\draw (1,0) circle (.5cm);
\draw (3,0) ellipse (1cm and .5cm) -- ++(3,0) circle (.5cm)
-- ++(2,-.5) circle (.25cm);
\end{tikzpicture}
11.8 The Arc Operation
The arc operation allows you to add an arc to the current path.
\path . . . arc(¸start angle):¸end angle):¸radius) and ¸half height)) . . . ;
The arc operation adds a part of a circle of the given radius between the given angles. The arc will
start at the current point and will end at the end of the arc.
97
\begin{tikzpicture}
\draw (0,0) arc (180:90:1cm) -- (2,.5) arc (90:0:1cm);
\draw (4,0) -- +(30:1cm) arc (30:60:1cm) -- cycle;
\draw (8,0) arc (0:270:1cm and .5cm) -- cycle;
\end{tikzpicture}
α
β
\begin{tikzpicture}
\draw (-1,0) -- +(3.5,0);
\draw (1,0) ++(210:2cm) -- +(30:4cm);
\draw (1,0) +(0:1cm) arc (0:30:1cm);
\draw (1,0) +(180:1cm) arc (180:210:1cm);
\path (1,0) ++(15:.75cm) node{$\alpha$};
\path (1,0) ++(15:-.75cm) node{$\beta$};
\end{tikzpicture}
11.9 The Grid Operation
You can add a grid to the current path using the grid path operation.
\path . . . grid[¸options)]¸corner) . . . ;
This operations adss a grid filling a rectangle whose two corners are given by ¸corner) and by the previous
coordinate. Thus, the typical way in which a grid is drawn is \draw (1,1) grid (3,3);, which yields
a grid filling the rectangle whose corners are at (1, 1) and (3, 3). All coordinate transformations apply
to the grid.
\tikz[rotate=30] \draw[step=1mm] (0,0) grid (2,2);
The ¸options), which are local to the grid operation, can be used to influence the appearance of the
grid. The stepping of the grid is governed by the following options:
• step=¸number or dimension or coordinate) sets the stepping in both the x and y-direction. If a
dimension is provided, this is used directly. If a number is provided, this number is interpreted in
the xy-coordinate system. For example, if you provide the number 2, then the x-step is twice the
x-vector and the y-step is twice the y-vector set by the x= and y= options. Finally, if you provide
a coordinate, then the x-part of this coordinate will be used as the x-step and the y-part will be
used as the y-coordinate.
\begin{tikzpicture}[x=.5cm]
\draw[thick] (0,0) grid [step=1] (3,2);
\draw[red] (0,0) grid [step=.75cm] (3,2);
\end{tikzpicture}
\begin{tikzpicture}
\draw (0,0) circle (1);
\draw[blue] (0,0) grid [step=(45:1)] (3,2);
\end{tikzpicture}
A complication arises when the x- and/or y-vector do not point along the axes. Because
of this, the actual rule for computing the x-step and the y-step is the following: As the
98
x- and y-steps we use the x- and y-components or the following two vectors: The first vec-
tor is either (¸x-grid-step-number), 0) or (¸x-grid-step-dimension), 0pt), the second vector is
(0, ¸y-grid-step-number)) or (0pt, ¸x-grid-step-dimension)).
• xstep=¸dimension or number) sets the stepping in the x-direction.
\tikz \draw (0,0) grid [xstep=.5,ystep=.75] (3,2);
• ystep=¸dimension or number) sets the stepping in the y-direction.
It is important to note that the grid is always “phased” such that it contains the point (0, 0) if that point
happens to be inside the rectangle. Thus, the grid does not always have an intersection at the corner
points; this occurs only if the corner points are multiples of the stepping. Note that due to rounding
errors, the “last” lines of a grid may be omitted. In this case, you have to add an epsilon to the corner
points.
The following style is useful for drawing grids:
• style=help lines This style makes lines “subdued” by using thin gray lines for them. However,
this style is not installed automatically and you have to say for example:
\tikz \draw[style=help lines] (0,0) grid (3,3);
11.10 The Parabola Operation
The parabola path operation continues the current path with a parabola. A parabola is a (shifted and
scaled) curve defined by the equation f(x) = x
2
and looks like this: .
\path . . . parabola[¸options)]bend¸bend coordinate)¸coordinate) . . . ;
This operation adds a parabola through the current point and the given ¸coordinate). If the bend is
given, it specifies where the bend should go; the ¸options) can also be used to specify where the bend
is. By default, the bend is at the old current point.
\begin{tikzpicture}
\draw (0,0) rectangle (1,1.5)
(0,0) parabola (1,1.5);
\draw[xshift=1.5cm] (0,0) rectangle (1,1.5)
(0,0) parabola[bend at end] (1,1.5);
\draw[xshift=3cm] (0,0) rectangle (1,1.5)
(0,0) parabola bend (.75,1.75) (1,1.5);
\end{tikzpicture}
The following options influence parabolas:
• bend=¸coordinate) Has the same effect as saying bend¸coordinate) outside the ¸options). The option
specifies that the bend of the parabola should be at the given ¸coordinate). You have to take care
yourself that the bend position is a “valid” position; which means that if there is no parabola of
the form f(x) = ax
2
+bx+c that goes through the old current point, the given bend, and the new
current point, the result will not be a parabola.
There is one special property of the ¸coordinate): When a relative coordinate is given like +(0,0),
the position relative to which this coordinate is “flexible.” More precisely, this position lies some-
where on a line from the old current point to the new current point. The exact position depends
on the next option.
99
• bend pos=¸fraction) Specifies where the “previous” point is relative to which the bend is calculated.
The previous point will be at the ¸fraction)th part of the line from the old current point to the
new current point.
The idea is the following: If you say bend pos=0 and bend +(0,0), the bend will be at the old
current point. If you say bend pos=1 and bend +(0,0), the bend will be at the new current point.
If you say bend pos=0.5 and bend +(0,2cm) the bend will be 2cm above the middle of the line
between the start and end point. This is most useful in situations such as the following:
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\draw (-1,0) parabola[bend pos=0.5] bend +(0,2) +(3,0);
\end{tikzpicture}
In the above example, the bend +(0,2) essentially means “a parabola that is 2cm high” and
+(3,0) means “and 3cm wide.” Since this situation arises often, there is a special shortcut option:
• parabola height=¸dimension) This option has the same effect as if you had written the following
instead: [bend pos=0.5,bend={+(0pt,¸dimension))}].
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\draw (-1,0) parabola[parabola height=2cm] +(3,0);
\end{tikzpicture}
The following styles are useful shortcuts:
• style=bend at start This places the bend at the start of a parabola. It is a shortcut for the
following options: bend pos=0,bend={+(0,0)}.
• style=bend at end This places the bend at the end of a parabola.
11.11 The Sine and Cosine Operation
The sin and cos operations are similar to the parabola operation. They, too, can be used to draw (parts
of) a sine or cosine curve.
\path . . . sin¸coordinate) . . . ;
The effect of sin is to draw a scaled and shifted version of a sine curve in the interval [0, π/2]. The
scaling and shifting is done in such a way that the start of the sine curve in the interval is at the old
current point and that the end of the curve in the interval is at ¸coordinate). Here is an example that
should clarify this:
\tikz \draw (0,0) rectangle (1,1) (0,0) sin (1,1)
(2,0) rectangle +(1.57,1) (2,0) sin +(1.57,1);
\path . . . cos¸coordinate) . . . ;
This operation works similarly, only a cosine in the interval [0, π/2] is drawn. By correctly alternating
sin and cos operations, you can create a complete sine or cosine curve:
100
\begin{tikzpicture}[xscale=1.57]
\draw (0,0) sin (1,1) cos (2,0) sin (3,-1) cos (4,0) sin (5,1);
\draw[color=red] (0,1.5) cos (1,0) sin (2,-1.5) cos (3,0) sin (4,1.5) cos (5,0);
\end{tikzpicture}
Note that there is no way to (conveniently) draw an interval on a sine or cosine curve whose end points
are not multiples of π/2.
11.12 The Plot Operation
The plot operation can be used to append a line or curve to the path that goes through a large number of
coordinates. These coordinates are either given in a simple list of coordinates, read from some file, or they
are computed on the fly.
Since the syntax and the behaviour of this command are a bit complex, they are described in the separated
Section 16.
11.13 The To Path Operation
The to operation is used to add a user-defined path from the previous coordinate to the following coordinate.
When you write (a) to (b), a straight line is added from a to b, exactly as if you had written (a) -- (b).
However, if you write (a) to [out=135,in=45] (b) a curve is added to the path, which leaves at an angle
of 135

at a and arrives at an angle of 45

at b. This is because the options in and out trigger a special
path to be used instead of the straight line.
\path . . . to[¸options)] ¸nodes) (¸coordinate)) . . . ;
This path operation inserts the path current set via the to path option at the current position. The
¸options) can be used to modify (perhaps implicitly) the to path and to setup how the path will be
rendered.
Before the to path is inserted, a number of macros are setup that can “help” the to path. These are
\tikztostart, \tikztotarget, and \tikztonodes; they are explained in the following.
Start and Target Coordinates. The to operation is always followed by a ¸coordinate), called the
target coordinate. The macro \tikztotarget is set to this coordinate (without the parantheses).
There is also a start coordinate, which is the coordinate preceding the to operation. This coordinate
can be accessed via the macro \tikztostart. In the following example, for the first to, the macro
\tikztostart is 0pt,0pt and the \tikztotarget is 0,2. For the second to, the macro \tikztostart
is 10pt,10pt and \tikztotarget is a.
a
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\draw (0,0) to (0,2);
\node (a) at (2,2) {a};
\draw[red] (10pt,10pt) to (a);
\end{tikzpicture}
Nodes on tos. It is possible to add nodes to the paths constructed by a to operation. To do so, you
specify the nodes between the to keyword and the coordinate (if there are options to the to operation,
these come first). The effect of (a) to node {x} (b) (typically) is the same as if you had written
(a) -- node {x} (b), namely that the node is placed on the to. This can be used to add labels to tos:
x
x
\begin{tikzpicture}
\draw (0,0) to node [sloped,above] {x} (3,2);
\draw (0,0) to[out=90,in=180] node [sloped,above] {x} (3,2);
\end{tikzpicture}
Styles for nodes. In addition to the ¸options) given after the to operation, the following style is also
set at the beginning of the to path:
101
• style=every to This style is installed at the beginning of every to. By default, it is set to draw.
\begin{tikzpicture}
\tikzstyle{every to}=[draw,dashed]
\path (0,0) to (3,2);
\end{tikzpicture}
Options. The ¸options) given with the to allow you to influence the appearance of the to path.
Mostly, these options are used to change the to path. This can be used to change the path from a
straight line to, say, a curve.
The path used is set using the following option:
• to path=¸path) Whenever an to operation is used, the ¸path) is inserted. More precisely, the
following path is added:
[every to,¸options)] ¸path)
The ¸options) are the options given to the to operation, the ¸path) is the path set by this option
to path.
Inside the ¸path), different macros are used to reference the from- and to-coordinates. In detail,
these are:
– \tikztostart will expand to the from-coordinate (without the parantheses).
– \tikztotarget will expand to the to-coordinate.
– \tikztonodes will expand to the nodes between the to operation and the coordinate. Fur-
thermore, these nodes will have the pos option set implicitly.
Let us have a look at a simple example. The standard straight line for an to is achieved by the
following ¸path):
-- (\tikztotarget) \tikztonodes
Indeed, this is the default setting for the path. When we write (a) to (b), the ¸path) will expand
to (a) -- (b), when we write
(a) to[red] node {x} (b)
the ¸path) will expand to
(a) -- (b) node[pos] {x}
It is not possible to specify the path
-- \tikztonodes (\tikztotarget)
since TikZ does not allow one to have a macro after -- that expands to a node.
Now let us have a look at how we can modify the ¸path) sensibly. The simplest way is to use a
curve.
a
b
c
x
\begin{tikzpicture}[to path={
.. controls +(1,0) and +(1,0) .. (\tikztotarget) \tikztonodes}]
\node (a) at (0,0) {a};
\node (b) at (2,1) {b};
\node (c) at (1,2) {c};
\draw (a) to node {x} (b)
(a) to (c);
\end{tikzpicture}
Here is another example:
q
a
q
b
1
0
\tikzstyle{my loop}=[->,to path={
.. controls +(80:1) and +(100:1) .. (\tikztotarget) \tikztonodes}]
\tikzstyle{my state}= [circle,draw]
\begin{tikzpicture}[shorten >=2pt]
\node [my state] (a) at (210:1) {$q_a$};
\node [my state] (b) at (330:1) {$q_b$};
\draw (a) to node[below] {1} (b)
to [my loop] node[above right] {0} (b);
\end{tikzpicture}
102
• execute at begin to=¸code) The ¸code) is executed prior to the to. This can be used to draw
one or more additional paths or to do additional computations.
• executed at end to=¸code) Works like the previous option, only this code is executed after the
to path has been added.
• style=every to This style is installed at the beginning of every to. It is empty by default.
There are a number of predefined to paths, see Section 32 for a reference.
11.14 The Scoping Operation
When TikZ encounters and opening or a closing brace ({ or }) at some point where a path operation should
come, it will open or close a scope. All options that can be applied “locally” will be scoped inside the
scope. For example, if you apply a transformation like [xshift=1cm] inside the scoped area, the shifting
only applies to the scope. On the other hand, an option like color=red does not have any effect inside a
scope since it can only be applied to the path as a whole.
11.15 The Node Operation
There are teo more operations that can be found in paths: node and edge. The first is used to add a so-called
node to a path. This operation is special in the following sense: It does not change the current path in any
way. In other words, this operation is not really a path operation, but has an effect that is “external” to the
path. The edge operation has similar effect in that it adds something after the main parth has been drawn.
However, it works like the to operation, that is, it adds a to path to the picture after the main path has
been drawn.
Since these operations are quite complex, they are described in the separate Section 13.
11.16 The PGF-Extra Operation
In some cases you may need to “do some calculations or some other stuff” while a path is constructed. For
this, you would like to suspend the construction of the path and suspend TikZ’s parsing of the path, you
would then like to have some T
E
X code executed, and would then like to resume the parsing of the path.
This effect can be achieved using the following path operation \pgfextra. Note that this operation should
only be used by real experts and should only be used deep inside clever macros, not on normal paths.
\pgfextra{¸code)}
This command may only be used inside a TikZ path. There it is used like a normal path operation.
The construction of the path is temporarily suspended and the ¸code) is executed. Then, the path
construction is resumed.
\newdimen\mydim
\begin{tikzpicture}
\mydim=1cm
\draw (0pt,\mydim) \pgfextra{\mydim=2cm} -- (0pt,\mydim);
\end{tikzpicture}
\pgfextra¸code) \endpgfextra
This is an alternative syntax for the \pgfextra command. If the code following \pgfextra does not
start with a brace, the ¸code) is executed until \endpgfextra is encountered. What actually happens is
that \pgfextra that is not followed by a brace completely shuts down the TikZ parse and \endpgfextra
is a normal macro that restarts the parser.
\newdimen\mydim
\begin{tikzpicture}
\mydim=1cm
\draw (0pt,\mydim)
\pgfextra \mydim=2cm \endpgfextra -- (0pt,\mydim);
\end{tikzpicture}
103
12 Actions on Paths
Once a path has been constructed, different things can be done with it. It can be drawn (or stroked) with
a “pen,” it can be filled with a color or shading, it can be used for clipping subsequent drawing, it can be
used to specify the extend of the picture—or any combination of these actions at the same time.
To decide what is to be done with a path, two methods can be used. First, you can use a special-purpose
command like \draw to indicate that the path should be drawn. However, commands like \draw and \fill
are just abbreviations for special cases of the more general method: Here, the \path command is used to
specify the path. Then, options encountered on the path indicate what should be done with the path.
For example, \path (0,0) circle (1cm); means “This is a path consisting of a circle around the origin.
Do not do anything with it (throw it away).” However, if the option draw is encountered anywhere on the
path, the circle will be drawn. “Anywhere” is any point on the path where an option can be given, which is
everywhere where a path command like circle (1cm) or rectangle (1,1) or even just (0,0) would also
be allowed. Thus, the following commands all draw the same circle:
\path [draw] (0,0) circle (1cm);
\path (0,0) [draw] circle (1cm);
\path (0,0) circle (1cm) [draw];
Finally, \draw (0,0) circle (1cm); also draws a path, because \draw is an abbreviation for
\path [draw] and thus the command expands to the first line of the above example.
Similarly, \fill is an abbreviation for \path[fill] and \filldraw is an abbreviation for the command
\path[fill,draw]. Since options accumulate, the following commands all have the same effect:
\path [draw,fill] (0,0) circle (1cm);
\path [draw] [fill] (0,0) circle (1cm);
\path [fill] (0,0) circle (1cm) [draw];
\draw [fill] (0,0) circle (1cm);
\fill (0,0) [draw] circle (1cm);
\filldraw (0,0) circle (1cm);
In the following subsection the different actions are explained that can be performed on a path. The
following commands are abbreviations for certain sets of actions, but for many useful combinations there are
no abbreviations:
\draw
Inside {tikzpicture} this is an abbreviation for \path[draw].
\fill
Inside {tikzpicture} this is an abbreviation for \path[fill].
\filldraw
Inside {tikzpicture} this is an abbreviation for \path[fill,draw].
\pattern
Inside {tikzpicture} this is an abbreviation for \path[pattern].
\shade
Inside {tikzpicture} this is an abbreviation for \path[shade].
\shadedraw
Inside {tikzpicture} this is an abbreviation for \path[shade,draw].
\clip
Inside {tikzpicture} this is an abbreviation for \path[clip].
\useasboundingbox
Inside {tikzpicture} this is an abbreviation for \path[use as bounding box].
104
12.1 Specifying a Color
The most unspecific option for setting colors is the following:
• color=¸color name)This option sets the color that is used for fill, drawing, and text inside the current
scope. Any special settings for filling colors or drawing colors are immediately “overruled” by this
option.
The ¸color name) is the name of a previously defined color. For L
A
T
E
X users, this is just a normal
“L
A
T
E
X-color” and the xcolor extensions are allows. Here is an example:
\tikz \fill[color=red!20] (0,0) circle (1ex);
It is possible to “leave out” the color= part and you can also write:
\tikz \fill[red!20] (0,0) circle (1ex);
What happens is that every option that TikZ does not know, like red!20, gets a “second chance” as
a color name.
For plain T
E
X users, it is not so easy to specify colors since plain T
E
X has no “standardized” color
naming mechanism. Because of this, pgf emulates the xcolor package, though the emulation is
extremely basic (more precisely, what I could hack together in two hours or so). The emulation allows
you to do the following:
– Specify a new color using \definecolor. Only the two color models gray and rgb are supported.
Example: \definecolor{orange}{rgb}{1,0.5,0}
– Use \colorlet to define a new color based on an old one. Here, the ! mechanism is supported,
though only “once” (use multiple \colorlet for more fancy colors).
Example: \colorlet{lightgray}{black!25}
– Use \color{¸color name)} to set the color in the current T
E
X group. \aftergroup-hackery is
used to restore the color after the group.
As pointed out above, the color= option applies to “everything” (except to shadings), which is not
always what you want. Because of this, there are several more specialized color options. For example, the
draw= option sets the color used for drawing, but does not modify the color used for filling. These color
options are documented where the path action they influence is described.
12.2 Drawing a Path
You can draw a path using the following option:
• draw=¸color) Causes the path to be drawn. “Drawing” (also known as “stroking”) can be thought of
as picking up a pen and moving it along the path, thereby leaving “ink” on the canvas.
There are numerous parameters that influence how a line is drawn, like the thickness or the dash
pattern. These options are explained below.
If the optional ¸color) argument is given, drawing is done using the given ¸color). This color can be
different from the current filling color, which allows you to draw and fill a path with different colors.
If no ¸color) argument is given, the last usage of the color= option is used.
If the special color name none is given, this option causes drawing to be “switched off.” This is useful
if a style has previously switched on drawing and you locally wish to undo this effect.
Although this option is normally used on paths to indicate that the path should be drawn, it also
makes sense to use the option with a {scope} or {tikzpicture} environment. However, this will not
cause all path to drawn. Instead, this just sets the ¸color) to be used for drawing paths inside the
environment.
\begin{tikzpicture}
\path[draw=red] (0,0) -- (1,1) -- (2,1) circle (10pt);
\end{tikzpicture}
The following subsections list the different options that influence how a path is drawn. All of these
options only have an effect if the draw options is given (directly or indirectly).
105
12.2.1 Graphic Parameters: Line Width, Line Cap, and Line Join
• line width=¸dimension) Specifies the line width. Note the space. Default: 0.4pt.
\tikz \draw[line width=5pt] (0,0) -- (1cm,1.5ex);
There are a number of predefined styles that provide more “natural” ways of setting the line width. You
can also redefine these styles. Remember that you can leave out the style= when setting a style.
• style=ultra thin Sets the line width to 0.1pt.
\tikz \draw[ultra thin] (0,0) -- (1cm,1.5ex);
• style=very thin Sets the line width to 0.2pt.
\tikz \draw[very thin] (0,0) -- (1cm,1.5ex);
• style=thin Sets the line width to 0.4pt.
\tikz \draw[thin] (0,0) -- (1cm,1.5ex);
• style=semithick Sets the line width to 0.6pt.
\tikz \draw[semithick] (0,0) -- (1cm,1.5ex);
• style=thick Sets the line width to 0.8pt.
\tikz \draw[thick] (0,0) -- (1cm,1.5ex);
• style=very thick Sets the line width to 1.2pt.
\tikz \draw[very thick] (0,0) -- (1cm,1.5ex);
• style=ultra thick Sets the line width to 1.6pt.
\tikz \draw[ultra thick] (0,0) -- (1cm,1.5ex);
• cap=¸type) Specifies how lines “end.” Permissible ¸type) are round, rect, and butt (default). They
have the following effects:
\begin{tikzpicture}
\begin{scope}[line width=10pt]
\draw[cap=rect] (0,0 ) -- (1,0);
\draw[cap=butt] (0,.5) -- (1,.5);
\draw[cap=round] (0,1 ) -- (1,1);
\end{scope}
\draw[white,line width=1pt]
(0,0 ) -- (1,0) (0,.5) -- (1,.5) (0,1 ) -- (1,1);
\end{tikzpicture}
• join=¸type) Specifies how lines “join.” Permissible ¸type) are round, bevel, and miter (default). They
have the following effects:
\begin{tikzpicture}[line width=10pt]
\draw[join=round] (0,0) -- ++(.5,1) -- ++(.5,-1);
\draw[join=bevel] (1.25,0) -- ++(.5,1) -- ++(.5,-1);
\draw[join=miter] (2.5,0) -- ++(.5,1) -- ++(.5,-1);
\useasboundingbox (0,1.5); % make bounding box bigger
\end{tikzpicture}
106
• miter limit=¸factor) When you use the miter join and there is a very sharp corner (a small angle),
the miter join may protrude very far over the actual joining point. In this case, if it were to protrude
by more than ¸factor) times the line width, the miter join is replaced by a bevel join. Default value is
10.
\begin{tikzpicture}[line width=5pt]
\draw (0,0) -- ++(5,.5) -- ++(-5,.5);
\draw[miter limit=25] (6,0) -- ++(5,.5) -- ++(-5,.5);
\useasboundingbox (14,0); % make bounding box bigger
\end{tikzpicture}
12.2.2 Graphic Parameters: Dash Pattern
• dash pattern=¸dash pattern) Sets the dashing pattern. The syntax is the same as in metafont. For
example on 2pt off 3pt on 4pt off 4pt means “draw 2pt, then leave out 3pt, then draw 4pt once
more, then leave out 4pt again, repeat”.
\begin{tikzpicture}[dash pattern=on 2pt off 3pt on 4pt off 4pt]
\draw (0pt,0pt) -- (3.5cm,0pt);
\end{tikzpicture}
• dash phase=¸dash phase) Shifts the start of the dash pattern by ¸phase).
\begin{tikzpicture}[dash pattern=on 20pt off 10pt]
\draw[dash phase=0pt] (0pt,3pt) -- (3.5cm,3pt);
\draw[dash phase=10pt] (0pt,0pt) -- (3.5cm,0pt);
\end{tikzpicture}
As for the line thickness, some predefined styles allow you to set the dashing conveniently.
• style=solid Shorthand for setting a solid line as “dash pattern.” This is the default.
\tikz \draw[solid] (0pt,0pt) -- (50pt,0pt);
• style=dotted Shorthand for setting a dotted dash pattern.
\tikz \draw[dotted] (0pt,0pt) -- (50pt,0pt);
• style=densely dotted Shorthand for setting a densely dotted dash pattern.
\tikz \draw[densely dotted] (0pt,0pt) -- (50pt,0pt);
• style=loosely dotted Shorthand for setting a loosely dotted dash pattern.
\tikz \draw[loosely dotted] (0pt,0pt) -- (50pt,0pt);
• style=dashed Shorthand for setting a dashed dash pattern.
\tikz \draw[dashed] (0pt,0pt) -- (50pt,0pt);
• style=densely dashed Shorthand for setting a densely dashed dash pattern.
\tikz \draw[densely dashed] (0pt,0pt) -- (50pt,0pt);
• style=loosely dashed Shorthand for setting a loosely dashed dash pattern.
\tikz \draw[loosely dashed] (0pt,0pt) -- (50pt,0pt);
107
12.2.3 Graphic Parameters: Draw Opacity
When a line is drawn, it will normally “obscure” everything behind it as if you has used perfectly opaque
ink. It is also possible to ask TikZ to use an ink that is a little bit (or a big bit) transparent. To do so, use
the following option:
• draw opacity=¸value) This option sets “how transparent” lines should be. A value of 1 means “fully
opaque” or “not transparent at all,” a value of 0 means “fully transparent” or “invisible.” A value of
0.5 yields lines that are semitransparent.
Note that when you use PostScript as your output format, this option works only with recent versions
of GhostScript.
\begin{tikzpicture}[line width=1ex]
\draw (0,0) -- (3,1);
\filldraw [fill=examplefill,draw opacity=0.5] (1,0) rectangle (2,1);
\end{tikzpicture}
Note that the draw opacity options only sets the opacity of drawn lines. The opacity of fillings is set
using the option fill opacity (documented in Section 12.3.3. The option opacity sets both at the same
time.
• opacity=¸value) Sets both the drawing and filling opacity to ¸value).
The following predefined styles make it easier to use this option:
– style=transparent Makes everything totally transparent and, hence, invisible.
\tikz{\fill[red] (0,0) rectangle (1,0.5);
\fill[transparent,red] (0.5,0) rectangle (1.5,0.25); }
– style=ultra nearly transparent Makes everything, well, ultra nearly transparent.
\tikz{\fill[red] (0,0) rectangle (1,0.5);
\fill[ultra nearly transparent] (0.5,0) rectangle (1.5,0.25); }
– style=very nearly transparent
\tikz{\fill[red] (0,0) rectangle (1,0.5);
\fill[very nearly transparent] (0.5,0) rectangle (1.5,0.25); }
– style=nearly transparent
\tikz{\fill[red] (0,0) rectangle (1,0.5);
\fill[nearly transparent] (0.5,0) rectangle (1.5,0.25); }
– style=semitransparent
\tikz{\fill[red] (0,0) rectangle (1,0.5);
\fill[semitransparent] (0.5,0) rectangle (1.5,0.25); }
– style=nearly opaque
\tikz{\fill[red] (0,0) rectangle (1,0.5);
\fill[nearly opaque] (0.5,0) rectangle (1.5,0.25); }
– style=very nearly opaque
\tikz{\fill[red] (0,0) rectangle (1,0.5);
\fill[very nearly opaque] (0.5,0) rectangle (1.5,0.25); }
– style=ultra nearly opaque
\tikz{\fill[red] (0,0) rectangle (1,0.5);
\fill[ultra nearly opaque] (0.5,0) rectangle (1.5,0.25); }
– style=opaque This yields completely opaque drawings, which is the default.
\tikz{\fill[red] (0,0) rectangle (1,0.5);
\fill[opaque] (0.5,0) rectangle (1.5,0.25); }
108
12.2.4 Graphic Parameters: Arrow Tips
When you draw a line, you can add arrow tips at the ends. It is only possible to add one arrow tip at the
start and one at the end. If the path consists of several segments, only the last segment gets arrow tips. The
behavior for paths that are closed is not specified and may change in the future.
• arrows=¸start arrow kind)-¸end arrow kind) This option sets the start and end arrow tips (an empty
value as in -> indicates that no arrow tip should be drawn at the start).
Note: Since the arrow option is so often used, you can leave out the text arrows=. What happens is
that every option that contains a - is interpreted as an arrow specification.
\begin{tikzpicture}
\draw[->] (0,0) -- (1,0);
\draw[o-stealth] (0,0.3) -- (1,0.3);
\end{tikzpicture}
The permissible values are all predefined arrow tips, though you can also define new arrow tip kinds
as explained in Section 48. This is often necessary to obtain “double” arrow tips and arrow tips that
have a fixed size. Since pgflibraryarrows is loaded by default, all arrow tips described in Section 18
are available.
One arrow tip kind is special: > (and all arrow tip kinds containing the arrow tip kind such as << or
>|). This arrow tip type is not fixed. Rather, you can redefine it using the >= option, see below.
Example: You can also combine arrow tip types as in
\begin{tikzpicture}[thick]
\draw[to reversed-to] (0,0) .. controls +(.5,0) and +(-.5,-.5) .. +(1.5,1);
\draw[[-latex reversed] (1,0) .. controls +(.5,0) and +(-.5,-.5) .. +(1.5,1);
\draw[latex-)] (2,0) .. controls +(.5,0) and +(-.5,-.5) .. +(1.5,1);
\useasboundingbox (-.1,-.1) rectangle (3.1,1.1); % make bounding box bigger
\end{tikzpicture}
• >=¸end arrow kind) This option can be used to redefine the “standard” arrow tip >. The idea is that
different people have different ideas what arrow tip kind should normally be used. I prefer the arrow
tip of T
E
X’s \to command (which is used in things like f : A → B). Other people will prefer L
A
T
E
X’s
standard arrow tip, which looks like this: . Since the arrow tip kind > is certainly the most “natural”
one to use, it is kept free of any predefined meaning. Instead, you can change it by saying >=to to set
the “standard” arrow tip kind to T
E
X’s arrow tip, whereas >=latex will set it to L
A
T
E
X’s arrow tip
and >=stealth will use a pstricks-like arrow tip.
Apart from redefining the arrow tip kind > (and < for the start), this option also redefines the following
arrow tip kinds: > and < as the swapped version of ¸end arrow kind), << and >> as doubled versions,
>> and << as swapped doubled versions, and |< and >| as arrow tips ending with a vertical bar.
\begin{tikzpicture}[scale=2]
\begin{scope}[>=latex]
\draw[->] (0pt,6ex) -- (1cm,6ex);
\draw[>->>] (0pt,5ex) -- (1cm,5ex);
\draw[|<->|] (0pt,4ex) -- (1cm,4ex);
\end{scope}
\begin{scope}[>=diamond]
\draw[->] (0pt,2ex) -- (1cm,2ex);
\draw[>->>] (0pt,1ex) -- (1cm,1ex);
\draw[|<->|] (0pt,0ex) -- (1cm,0ex);
\end{scope}
\end{tikzpicture}
• shorten >=¸dimension) This option will shorten the end of lines by the given ¸dimension). If you
specify an arrow tip, lines are already shortened a bit such that the arrow tip touches the specified
endpoint and does not “protrude over” this point. Here is an example:
109
\begin{tikzpicture}[line width=20pt]
\useasboundingbox (0,-1.5) rectangle (3.5,1.5);
\draw[red] (0,0) -- (3,0);
\draw[gray,->] (0,0) -- (3,0);
\end{tikzpicture}
The shorten > option allows you to shorten the end on the line additionally by the given distance.
This option can also be useful if you have not specified an arrow tip at all.
\begin{tikzpicture}[line width=20pt]
\useasboundingbox (0,-1.5) rectangle (3.5,1.5);
\draw[red] (0,0) -- (3,0);
\draw[-to,shorten >=10pt,gray] (0,0) -- (3,0);
\end{tikzpicture}
• shorten <=¸dimension) works like shorten >, but for the start.
12.2.5 Graphic Parameters: Double Lines and Bordered Lines
• double=¸core color) This option causes “two” lines to be drawn instead of a single one. However, this
is not what really happens. In reality, the path is drawn twice. First, with the normal drawing color,
secondly with the ¸core color), which is normally white. Upon the second drawing, the line width is
reduced. The net effect is that it appears as if two lines had been drawn and this works well even with
complicated, curved paths:
\tikz \draw[double]
plot[smooth cycle] coordinates{(0,0) (1,1) (1,0) (0,1)};
You can also use the doubling option to create an effect in which a line seems to have a certain “border”:
\begin{tikzpicture}
\draw (0,0) -- (1,1);
\draw[draw=white,double=red,very thick] (0,1) -- (1,0);
\end{tikzpicture}
• double distance=¸dimension) Sets the distance the “two” lines are spaced apart (default is 0.6pt).
In reality, this is the thickness of the line that is used to draw the path for the second time. The
thickness of the first time the path is drawn is twice the normal line width plus the given ¸dimension).
As a side-effect, this option “selects” the double option.
\begin{tikzpicture}
\draw[very thick,double] (0,0) arc (180:90:1cm);
\draw[very thick,double distance=2pt] (1,0) arc (180:90:1cm);
\draw[thin,double distance=2pt] (2,0) arc (180:90:1cm);
\end{tikzpicture}
12.3 Filling a Path
To fill a path, use the following option:
• fill=¸color) This option causes the path to be filled. All unclosed parts of the path are first closed,
if necessary. Then, the area enclosed by the path is filled with the current filling color, which is either
the last color set using the general color= option or the optional color ¸color). For self-intersection
paths and for paths consisting of several closed areas, the “enclosed area” is somewhat complicated to
110
define and two different definitions exist, namely the nonzero winding number rule and the even odd
rule, see the explanation of these options, below.
Just as for the draw option, setting ¸color) to none disables filling locally.
\begin{tikzpicture}
\fill (0,0) -- (1,1) -- (2,1);
\fill (4,0) circle (.5cm) (4.5,0) circle (.5cm);
\fill[even odd rule] (6,0) circle (.5cm) (6.5,0) circle (.5cm);
\fill (8,0) -- (9,1) -- (10,0) circle (.5cm);
\end{tikzpicture}
If the fill option is used together with the draw option (either because both are given as options or
because a \filldraw command is used), the path is filled first, then the path is drawn second. This
is especially useful if different colors are selected for drawing and for filling. Even if the same color is
used, there is a difference between this command and a plain fill: A “filldrawn” area will be slightly
larger than a filled area because of the thickness of the “pen.”
\begin{tikzpicture}[fill=examplefill,line width=5pt]
\filldraw (0,0) -- (1,1) -- (2,1);
\filldraw (4,0) circle (.5cm) (4.5,0) circle (.5cm);
\filldraw[even odd rule] (6,0) circle (.5cm) (6.5,0) circle (.5cm);
\filldraw (8,0) -- (9,1) -- (10,0) circle (.5cm);
\end{tikzpicture}
12.3.1 Graphic Parameters: Fill Pattern
Instead of filling a path with a single solid color, it is also possible to fill it with a tiling pattern. Imagine a
small tile that contains a simple picture like a star. Then these tiles are (conceptually) repeated infinitely
in all directions, but clipped against the path.
Tiling patterns come in two variants: inherently colored patterns and form-only patterns. An inherently
colored pattern is, say, a red star with a black border and will always look like this. A form-only pattern
may have a different color each time it is used, only the form of the pattern will stay the same. As such,
form-only patterhns do not have any colors of their own, but when it is used the current pattern color is
used as its color.
Patterns are not overly flexible. In particular, it is not possible to change the size or orientation of a
pattern without declaring a new pattern. For complicated case, it may be easier to use two nested \foreach
statements to simulate a pattern, but patterns are rendered much more quickly than simulated ones.
• pattern=¸name) This option causes the path to be filled with a pattern. If the ¸name) is given, this
pattern is used, otherwise the pattern set in the enclosing scope is used. As for the draw and fill
options, setting ¸name) to none disables filling locally.
The pattern works like a fill color. In particular, setting a new fill color will fill the path with a solid
color once more.
Strangely, no ¸name)s are permissible by default. You neet to load for instance pgflibrarypatterns,
see Section 26, to install predefined patterns.
\begin{tikzpicture}
\draw[pattern=dots] (0,0) circle (1cm);
\draw[pattern=fivepointed stars] (0,0) rectangle (3,1);
\end{tikzpicture}
111
• pattern color=¸color) This option is used to set the color to be used for form-only patterns. This
option has no effect on inherently colored patterns.
\begin{tikzpicture}
\draw[pattern color=red,pattern=fivepointed stars] (0,0) circle (1cm);
\draw[pattern color=blue,pattern=fivepointed stars] (0,0) rectangle (3,1);
\end{tikzpicture}
\begin{tikzpicture}
\def\mypath{(0,0) -- +(0,1) arc (180:0:1.5cm) -- +(0,-1)}
\fill [red] \mypath;
\pattern[pattern color=white,pattern=bricks] \mypath;
\end{tikzpicture}
12.3.2 Graphic Parameters: Interior Rules
The following two options can be used to decide how interior points should be determined:
• nonzero rule If this rule is used (which is the default), the following method is used to determine
whether a given point is “inside” the path: From the point, shoot a ray in some direction towards
infinity (the direction is chosen such that no strange borderline cases occur). Then the ray may hit
the path. Whenever it hits the path, we increase or decrease a counter, which is initially zero. If the
ray hits the path as the path goes “from left to right” (relative to the ray), the counter is increased,
otherwise it is decreased. Then, at the end, we check whether the counter is nonzero (hence the name).
If so, the point is deemed to lie “inside,” otherwise it is “outside.” Sounds complicated? It is.
crossings: −1 + 1 = 0
crossings: 1 + 1 = 2
\begin{tikzpicture}
\filldraw[fill=examplefill]
% Clockwise rectangle
(0,0) -- (0,1) -- (1,1) -- (1,0) -- cycle
% Counter-clockwise rectangle
(0.25,0.25) -- (0.75,0.25) -- (0.75,0.75) -- (0.25,0.75) -- cycle;
\draw[->] (0,1) -- (.4,1);
\draw[->] (0.75,0.75) -- (0.3,.75);
\draw[->] (0.5,0.5) -- +(0,1) node[above] {crossings: $-1+1 = 0$};
\begin{scope}[yshift=-3cm]
\filldraw[fill=examplefill]
% Clockwise rectangle
(0,0) -- (0,1) -- (1,1) -- (1,0) -- cycle
% Clockwise rectangle
(0.25,0.25) -- (0.25,0.75) -- (0.75,0.75) -- (0.75,0.25) -- cycle;
\draw[->] (0,1) -- (.4,1);
\draw[->] (0.25,0.75) -- (0.4,.75);
\draw[->] (0.5,0.5) -- +(0,1) node[above] {crossings: $1+1 = 2$};
\end{scope}
\end{tikzpicture}
• even odd rule This option causes a different method to be used for determining the inside and outside
of paths. While it is less flexible, it turns out to be more intuitive.
With this method, we also shoot rays from the point for which we wish to determine whether it is
inside or outside the filling area. However, this time we only count how often we “hit” the path and
declare the point to be “inside” if the number of hits is odd.
Using the even-odd rule, it is easy to “drill holes” into a path.
112
crossings: 1 + 1 = 2
\begin{tikzpicture}
\filldraw[fill=examplefill,even odd rule]
(0,0) rectangle (1,1) (0.5,0.5) circle (0.4cm);
\draw[->] (0.5,0.5) -- +(0,1) [above] node{crossings: $1+1 = 2$};
\end{tikzpicture}
12.3.3 Graphic Parameters: Fill Opacity
Analogously to the draw opacity, you can also set the filling opacity:
• fill opacity=¸value) This option sets the opacity of fillings. In addition to filling operations, this
opacity also applies to text and images.
Note, again, that when you use PostScript as your output format, this option works only with recent
versions of GhostScript.
\begin{tikzpicture}[thick,fill opacity=0.5]
\filldraw[fill=red] (0:1cm) circle (12mm);
\filldraw[fill=green] (120:1cm) circle (12mm);
\filldraw[fill=blue] (-120:1cm) circle (12mm);
\end{tikzpicture}
A
B
\begin{tikzpicture}
\fill[red] (0,0) rectangle (3,2);
\node at (0,0) {\huge A};
\node[fill opacity=0.5] at (3,2) {\huge B};
\end{tikzpicture}
12.4 Shading a Path
You can shade a path using the shade option. A shading is like a filling, only the shading changes its color
smoothly from one color to another.
• shade Causes the path to be shaded using the currently selected shading (more on this later). If this
option is used together with the draw option, then the path is first shaded, then drawn.
It is not an error to use this option together with the fill option, but it makes no sense.
\tikz \shade (0,0) circle (1ex);
\tikz \shadedraw (0,0) circle (1ex);
For some shadings it is not really clear how they can “fill” the path. For example, the ball shading
normally looks like this: . How is this supposed to shade a rectangle? Or a triangle?
To solve this problem, the predefined shadings like ball or axis fill a large rectangle completely in a
sensible way. Then, when the shading is used to “shade” a path, what actually happens is that the path
is temporarily used for clipping and then the rectangular shading is drawn, scaled and shifted such that all
parts of the path are filled.
113
12.4.1 Choosing a Shading Type
The default shading is a smooth transition from gray to white and from above to bottom. However, other
shadings are also possible, for example a shading that will sweep a color from the center to the corners
outward. To choose the shading, you can use the shading= option, which will also automatically invoke the
shade option. Note that this does not change the shading color, only the way the colors sweep. For changing
the colors, other options are needed, which are explained below.
• shading=¸name) This selects a shading named ¸name). The following shadings are predefined:
– axis This is the default shading in which the color changes gradually between three horizontal
lines. The top line is at the top (uppermost) point of the path, the middle is in the middle, the
bottom line is at the bottom of the path.
\tikz \shadedraw [shading=axis] (0,0) rectangle (1,1);
The default top color is gray, the default bottom color is white, the default middle is the “middle”
of these two.
– radial This shading fills the path with a gradual sweep from a certain color in the middle to
another color at the border. If the path is a circle, the outer color will be reached exactly at the
border. If the shading is not a circle, the outer color will continue a bit towards the corners. The
default inner color is gray, the default outer color is white.
\tikz \shadedraw [shading=radial] (0,0) rectangle (1,1);
– ball This shading fills the path with a shading that “looks like a ball.” The default “color” of
the ball is blue (for no particular reason).
\tikz \shadedraw [shading=ball] (0,0) rectangle (1,1);
\tikz \shadedraw [shading=ball] (0,0) circle (.5cm);
• shading angle=¸degrees) This option rotates the shading (not the path!) by the given angle. For
example, we can turn a top-to-bottom axis shading into a left-to-right shading by rotating it by 90

.
\tikz \shadedraw [shading=axis,shading angle=90] (0,0) rectangle (1,1);
You can also define new shading types yourself. However, for this, you need to use the basic layer directly,
which is, well, more basic and harder to use. Details on how to create a shading appropriate for filling paths
are given in Section 55.3.
12.4.2 Choosing a Shading Color
The following options can be used to change the colors used for shadings. When one of these options is
given, the shade option is automatically selected and also the “right” shading.
• top color=¸color) This option prescribes the color to be used at the top in an axis shading. When
this option is given, several things happen:
1. The shade option is selected.
114
2. The shading=axis option is selected.
3. The middle color of the axis shading is set to the average of the given top color ¸color) and of
whatever color is currently selected for the bottom.
4. The rotation angle of the shading is set to 0.
\tikz \draw[top color=red] (0,0) rectangle (2,1);
• bottom color=¸color) This option works like top color, only for the bottom color.
• middle color=¸color) This option specifies the color for the middle of an axis shading. It also sets
the shade and shading=axis options, but it does not change the rotation angle.
Note: Since both top color and bottom color change the middle color, this option should be given
last if all of these options need to be given:
\tikz \draw[top color=white,bottom color=black,middle color=red]
(0,0) rectangle (2,1);
• left color=¸color) This option does exactly the same as top color, except that the shading angle
is set to 90

.
• right color=¸color) Works like left color.
• inner color=¸color) This option sets the color used at the center of a radial shading. When this
option is used, the shade and shading=radial options are set.
\tikz \draw[inner color=red] (0,0) rectangle (2,1);
• outer color=¸color) This option sets the color used at the border and outside of a radial shading.
\tikz \draw[outer color=red,inner color=white]
(0,0) rectangle (2,1);
• ball color=¸color) This option sets the color used for the ball shading. It sets the shade and
shading=ball options. Note that the ball will never “completely” have the color ¸color). At its
“highlight” spot a certain amount of white is mixed in, at the border a certain amount of black.
Because of this, it also makes sense to say ball color=white or ball color=black
\begin{tikzpicture}
\shade[ball color=white] (0,0) circle (2ex);
\shade[ball color=red] (1,0) circle (2ex);
\shade[ball color=black] (2,0) circle (2ex);
\end{tikzpicture}
12.5 Establishing a Bounding Box
pgf is reasonably good at keeping track of the size of your picture and reserving just the right amount of space
for it in the main document. However, in some cases you may want to say things like “do not count this for the
picture size” or “the picture is actually a little large.” For this you can use the option use as bounding box
or the command \useasboundingbox, which is just a shorthand for \path[use as bounding box].
115
• use as bounding box Normally, when this option is given on a path, the bounding box of the present
path is used to determine the size of the picture and the size of all subsequent paths are ignored.
However, if there were previous path operations that have already established a larger bounding box,
it will not be made smaller by this operation.
In a sense, use as bounding box has the same effect as clipping all subsequent drawing against the
current path—without actually doing the clipping, only making pgf treat everything as if it were
clipped.
The first application of this option is to have a {tikzpicture} overlap with the main text:
Left of picture right of picture.
Left of picture\begin{tikzpicture}
\draw[use as bounding box] (2,0) rectangle (3,1);
\draw (1,0) -- (4,.75);
\end{tikzpicture}right of picture.
In a second application this option can be used to get better control over the white space around the
picture:
Left of picture right of picture.
Left of picture
\begin{tikzpicture}
\useasboundingbox (0,0) rectangle (3,1);
\fill (.75,.25) circle (.5cm);
\end{tikzpicture}
right of picture.
Note: If this option is used on a path inside a T
E
X group (scope), the effect “lasts” only till the end
of the scope. Again, this behavior is the same as for clipping.
There is a node that allows you to get the size of the current bounding box. The current bounding box
node has the rectangle shape and its size is always the size of the current bounding box.
Similarly, the current path bounding box node has the rectangle hape and the size of the bounding
box of the current path.
\begin{tikzpicture}
\draw[red] (0,0) circle (2pt);
\draw[red] (2,1) circle (3pt);
\draw (current bounding box.south west) rectangle
(current bounding box.north east);
\draw[red] (3,-1) circle (4pt);
\draw[thick] (current bounding box.south west) rectangle
(current bounding box.north east);
\end{tikzpicture}
12.6 Using a Path For Clipping
To use a path for clipping, use the clip option.
• clip This option causes all subsequent drawings to be clipped against the current path and the size
of subsequent paths will not be important for the picture size. If you clip against a self-intersecting
path, the even-odd rule or the nonzero winding number rule is used to determine whether a point is
inside or outside the clipping region.
The clipping path is a graphic state parameter, so it will be reset at the end of the current scope.
Multiple clippings accumulate, that is, clipping is always done against the intersection of all clipping
116
areas that have been specified inside the current scopes. The only way of enlarging the clipping area
is to end a {scope}.
\begin{tikzpicture}
\draw[clip] (0,0) circle (1cm);
\fill[red] (1,0) circle (1cm);
\end{tikzpicture}
It is usually a very good idea to apply the clip option only to the first path command in a scope.
If you “only wish to clip” and do not wish to draw anything, you can use the \clip command, which
is a shorthand for \path[clip].
\begin{tikzpicture}
\clip (0,0) circle (1cm);
\fill[red] (1,0) circle (1cm);
\end{tikzpicture}
To keep clipping local, use {scope} environments as in the following example:
\begin{tikzpicture}
\draw (0,0) -- ( 0:1cm);
\draw (0,0) -- (10:1cm);
\draw (0,0) -- (20:1cm);
\draw (0,0) -- (30:1cm);
\begin{scope}[fill=red]
\fill[clip] (0.2,0.2) rectangle (0.5,0.5);
\draw (0,0) -- (40:1cm);
\draw (0,0) -- (50:1cm);
\draw (0,0) -- (60:1cm);
\end{scope}
\draw (0,0) -- (70:1cm);
\draw (0,0) -- (80:1cm);
\draw (0,0) -- (90:1cm);
\end{tikzpicture}
There is a slightly annoying catch: You cannot specify certain graphic options for the command used
for clipping. For example, in the above code we could not have moved the fill=red to the \fill
command. The reasons for this have to do with the internals of the pdf specification. You do not
want to know the details. It is best simply not to specify any options for these commands.
117
13 Nodes and Edges
13.1 Overview
In the present section, the usage of nodes in TikZ is explained. A node is typically a rectangle or circle or
another simple shape with some text on it.
Nodes are added to paths using the special path operation node. Nodes are not part of the path itself.
Rather, they are added to the picture after the path has been drawn.
In Section 13.2 the basic syntax of the node operation is explained, followed in Section 13.3 by the syntax
for multi-part nodes, which are nodes that contain several different text parts. After this, the different
options for the text in nodes are explained. In Section 13.5 the concept of anchors is introduced along
with their usage. In Section 13.6 the different ways transformations affect nodes are studied. Sections 13.7
and 13.8 are about placing nodes on or next to straight lines and curves. In Section 13.10 it is explained
how a node can be used as a “pseudo-coordinate.” Section 13.11 introduces the edge operation, which works
similar to the to operation and also similar to the node operation. Section 13.13 lists the predefined shapes.
Finally, Section 13.14 explains the special after node path options.
13.2 Nodes and Their Shapes
In the simplest case, a node is just some text that is placed at some coordinate. However, a node can also
have a border drawn around it or have a more complex background and foreground. Indeed, some nodes do
not have a text at all, but consist solely of the background. You can name nodes so that you can reference
their coordinates later in the same picture or, if certain precautions are taken as explained in Section 13.12,
also in different pictures.
There are no special T
E
X commands for adding a node to a picture; rather, there is path operation called
node for this. Nodes are created whenever TikZ encounters node or coordinate at a point on a path where
it would expect a normal path operation (like -- (1,1) or sin (1,1)). It is also possible to give node
specifications inside certain path operations as explained later.
The node operation is typically followed by some options, which apply only to the node. Then, you can
optionally name the node by providing a name in round braces. Lastly, for the node operation you must
provide some label text for the node in curly braces, while for the coordinate operation you may not. The
node is placed at the current position of the path after the path has been drawn. Thus, all nodes are drawn
“on top” of the path and retained until the path is complete. If there are several nodes on a path, they are
drawn on top of the path in the order they are encountered.
first node
second node
third node
\tikz \fill[fill=examplefill]
(0,0) node {first node}
-- (1,1) node {second node}
-- (0,2) node {third node};
The syntax for specifying nodes is the following:
\path . . . node[¸options)](¸name))at(¸coordinate)){¸text)} . . . ;
The effect of at is to place the node at the coordinate given after at and not, as would normally be the
case, at the last position. The at syntax is not available when a node is given inside a path operation
(it would not make any sense, there).
The (¸name)) is a name for later reference and it is optional. You may also add the option name=¸name)
to the ¸option) list; it has the same effect.
• name=¸node name) assigns a name to the node for later reference. Since this is a “high-level” name
(drivers never know of it), you can use spaces, number, letters, or whatever you like when naming
a node. Thus, you can name a node just 1 or perhaps start of chart or even y_1. Your node
name should not contain any punctuation like a dot, a comma, or a colon since these are used to
detect what kind of coordinate you mean when you reference a node.
• at=¸coordinate) is another way of specifying ath at coordinate.
The ¸options) is an optional list of options that apply only to the node and have no effect outside. The
other way round, most “outside” options also apply to the node, but not all. For example, the “outside”
118
rotation does not apply to nodes (unless some special options are used, sigh). Also, the outside path
action, like draw or fill, never applies to the node and must be given in the node (unless some special
other options are used, deep sigh).
As mentioned before, we can add a border and even a background to a node:
first node
second node
third node
\tikz \fill[fill=examplefill]
(0,0) node {first node}
-- (1,1) node[draw] {second node}
-- (0,2) node[fill=red!20,draw,double,rounded corners] {third node};
The “border” is actually just a special case of a much more general mechanism. Each node has a certain
shape which, by default, is a rectangle. However, we can also ask TikZ to use a circle shape instead or
an ellipse shape (you have to include pgflibraryshapes for the latter shape):
first node
second node
third node
\tikz \fill[fill=examplefill]
(0,0) node{first node}
-- (1,1) node[ellipse,draw] {second node}
-- (0,2) node[circle,fill=red!20] {third node};
In the future, there might be much more complicated shapes available such as, say, a shape for a resistor
or a shape for a uml class. Unfortunately, creating new shapes is a bit tricky and makes it necessary
to use the basic layer directly. Life is hard.
To select the shape of a node, the following option is used:
• shape=¸shape name) select the shape either of the current node or, when this option is not given
inside a node but somewhere outside, the shape of all nodes in the current scope.
Since this option is used often, you can leave out the shape=. When TikZ encounters an option like
circle that it does not know, it will, after everything else has failed, check whether this option is
the name of some shape. If so, that shape is selected as if you had said shape=¸shape name).
By default, the following shapes are available: rectangle, circle, coordinate, and, when the
package pgflibraryshapes is loaded, also ellipse. Details of these shapes, like their anchors and
size options, are discussed in Section 13.13.
The following styles influences how nodes are rendered:
• style=every node This style is installed at the beginning of every node.
A
B
\begin{tikzpicture}
\tikzstyle{every node}=[draw]
\draw (0,0) node {A} -- (1,1) node {B};
\end{tikzpicture}
• style=every ¸shape ) node These styles are installed at the beginning of a node of a given ¸shape).
For example, every rectangle node is used for rectangle nodes, and so on.
A
B
\begin{tikzpicture}
\tikzstyle{every rectangle node}=[draw]
\tikzstyle{every circle node}= [draw,double]
\draw (0,0) node[rectangle] {A} -- (1,1) node[circle] {B};
\end{tikzpicture}
There is a special syntax for specifying “light-weighed” nodes:
\path . . . coordinate[¸options)](¸name))at(¸coordinate)) . . . ;
119
This has the same effect as
node[shape=coordinate][¸options)](¸name))at(¸coordinate)){},
where the at part might be missing.
Since nodes are often the only path operation on paths, there are two special commands for creating
paths containing only a node:
\node
Inside {tikzpicture} this is an abbreviation for \path node.
\coordinate
Inside {tikzpicture} this is an abbreviation for \path coordinate.
13.3 Multi-Part Nodes
Most nodes just have a single simple text label. However, nodes of a more complicated shapes might be
made up from several node parts. For example, in automata theory a so-called Moore state has a state name,
drawn in the upper part of the state circle, and an output text, drawn in the lower part of the state circle.
These two parts are quite independent. Similarly, a uml class shape would have a name part, a method
part, and an attributes part. Different molecule shape might use parts for the different atoms to be drawn
at the different positions, and so on.
Both pgf and TikZ support such multipart nodes. On the lower level, pgf provides a system for
specifying that a shape consists of several parts. On the TikZ level, you specify the different node parts by
using the following command:
\nodepart{¸part name)}
This command can only be used inside the ¸text) argument of a node path operation. It works a little
bit like a \part command in L
A
T
E
X. It will stop the typesetting of whatever node part was typeset until
now and then start putting all following text into the node part named ¸part name)—until another
\partname is encountered or until the node ¸text) ends.
q
1
00
\begin{tikzpicture}
\node [circle split,draw,double,fill=red!20]
{
% No \nodepart has been used, yet. So, the following is put in the
% ‘‘text’’ node part by default.
$q_1$
\nodepart{lower} % Ok, end ‘‘text’’ part, start ‘‘output’’ part
$00$
}; % output part ended.
\end{tikzpicture}
You will have to lookup which parts are defined by a shape.
The following styles influences node parts:
• style=every ¸part name ) node part This style is installed at the beginning of every node part
named ¸part name).
q
1
00
\tikzstyle{every lower node part}=[red]
\tikz \node [circle split,draw] {$q_1$ \nodepart{lower} $00$};
13.4 Options for the Text in Nodes
The simplest option for the text in nodes is its color. Normally, this color is just the last color installed
using color=, possibly inherited from another scope. However, it is possible to specificly set the color used
for text using the following option:
• text=¸color) Sets the color to be used for text labels. A color= option will immediately override this
option.
120
red red red
\begin{tikzpicture}
\draw[red] (0,0) -- +(1,1) node[above] {red};
\draw[text=red] (1,0) -- +(1,1) node[above] {red};
\draw (2,0) -- +(1,1) node[above,red] {red};
\end{tikzpicture}
Just like the color itself, you may also wish to set the opacity of the text only. For this, use the following
option:
• text opacity=¸value) Sets the opacity of text labels.
Upper node
Lower node
\begin{tikzpicture}
\draw[line width=2mm,blue!50,cap=round] (0,0) grid (3,2);
\tikzstyle{every node}=[fill,draw]
\node[opacity=0.5] at (1.5,2) {Upper node};
\node[draw opacity=0.8,fill opacity=0.2,text opacity=1]
at (1.5,0) {Lower node};
\end{tikzpicture}
Next, you may wish to adjust the font used for the text. Use the following option for this:
• font=¸font commands) Sets the font used for text labels.
italic
\begin{tikzpicture}
\draw[font=\itshape] (1,0) -- +(1,1) node[above] {italic};
\end{tikzpicture}
A perhaps more useful example is the following:
state
output
\tikzstyle{every text node part}=[font=\itshape]
\tikzstyle{every lower node part}=[font=\footnotesize]
\tikz \node [circle split,draw] {state \nodepart{lower} output};
Normally, when a node is typeset, all the text you give in the braces is but in one long line (in an \hbox,
to be precise) and the node will become as wide as necessary.
You can change this behaviour using the following options. They allow you to limit the width of a node
(naturally, at the expense of its height).
• text width=¸dimension) This option will put the text of a node in a box of the given width (more
precisely, in a {minipage} of this width; for plain T
E
X a rudimentary “minipage emulation” is used).
If the node text is not as wide as ¸dimension), it will nevertheless be put in a box of this width. If it
is larger, line breaking will be done.
By default, when this option is given, a ragged right border will be used. This is sensible since,
typically, these boxes are narrow and justifying the text looks ugly.
This is a demon-
stration text for
showing how line
breaking works.
\tikz \draw (0,0) node[fill=examplefill,text width=3cm]
{This is a demonstration text for showing how line breaking works.};
• text justified causes the text to be justified instead of (right)ragged. Use this only with pretty
broad nodes.
This is a demon-
stration text for
showing how line
breaking works.
\tikz \draw (0,0) node[fill=examplefill,text width=3cm,text justified]
{This is a demonstration text for showing how line breaking works.};
121
In the above example, T
E
X complains (rightfully) about three very badly typeset lines. (For this
manual I asked T
E
X to stop complaining by using \hbadness=10000, but this is a foul deed, indeed.)
• text ragged causes the text to be typeset with a ragged right. This uses the original plain T
E
X
definition of a ragged right border, in which T
E
X will try to balance the right border as well as
possible. This is the default.
This is a demon-
stration text for
showing how line
breaking works.
\tikz \draw (0,0) node[fill=examplefill,text width=3cm,text ragged]
{This is a demonstration text for showing how line breaking works.};
• text badly ragged causes the right border to be ragged in the L
A
T
E
X-style, in which no balancing
occurs. This looks ugly, but it may be useful for very narrow boxes and when you wish to avoid
hyphenations.
This is a
demonstration text
for showing how
line breaking
works.
\tikz \draw (0,0) node[fill=examplefill,text width=3cm,text badly ragged]
{This is a demonstration text for showing how line breaking works.};
• text centered centers the text, but tries to balance the lines.
This is a demon-
stration text for
showing how line
breaking works.
\tikz \draw (0,0) node[fill=examplefill,text width=3cm,text centered]
{This is a demonstration text for showing how line breaking works.};
• text badly centered centers the text, without balancing the lines.
This is a
demonstration text
for showing how
line breaking
works.
\tikz \draw (0,0) node[fill=examplefill,text width=3cm,text badly centered]
{This is a demonstration text for showing how line breaking works.};
In addition to changing the width of nodes, you can also change the height of nodes. This can be done in
two ways: First, you can use the option minimum height, which ensures that the height of the whole node
is at least the given height (this option is described in more detail later). Second, you can use the option
text height, which sets the height of the text itself, more precisely, of the T
E
X text box of the text. Note
that the text height typically is not the height of the shape’s box: In addition to the text height, an
internal inner sep is added as extra space and the text depth is also taken into account.
I recommend using minimum size instead of text height except for special situations.
• text height=¸dimension) Sets the height of the text boxes in shapes. Thus, when you write something
like node {text}, the text is first typeset, resulting in some box of a certain height. This height is
then replaced by the height text height. The resulting box is then used to determine the size of the
shape, which will typically be larger. When you write text height= without specifying anything, the
“natural” size of the text box remains unchanged.
y y
\tikz \node[draw] {y};
\tikz \node[draw,text height=10pt] {y};
• text depth=¸dimension) This option works like text height, only for the depth of the text box. This
option is mostly useful when you need to ensure a uniform depth of text boxes that need to be aligned.
122
13.5 Placing Nodes Using Anchors
When you place a node at some coordinate, the node is centered on this coordinate by default. This is often
undesirable and it would be better to have the node to the right or above the actual coordinate.
pgf uses a so-called anchoring mechanism to give you a very fine control over the placement. The idea
is simple: Imaging a node of rectangular shape of a certain size. pgf defines numerous anchor positions in
the shape. For example to upper right corner is called, well, not “upper right anchor,” but the north east
anchor of the shape. The center of the shape has an anchor called center on top of it, and so on. Here are
some examples (a complete list is given in Section 13.13).
Big node
north north east north west
west east
base
Now, when you place a node at a certain coordinate, you can ask TikZ to place the node shifted around
in such a way that a certain anchor is at the coordinate. In the following example, we ask TikZ to shift the
first node such that its north east anchor is at coordinate (0,0) and that the west anchor of the second
node is at coordinate (1,1).
first node
second node
\tikz \draw (0,0) node[anchor=north east] {first node}
rectangle (1,1) node[anchor=west] {second node};
Since the default anchor is center, the default behaviour is to shift the node in such a way that it is
centered on the current position.
• anchor=¸anchor name) causes the node to be shifted such that it’s anchor ¸anchor name) lies on the
current coordinate.
The only anchor that is present in all shapes is center. However, most shapes will at least define
anchors in all “compass directions.” Furthermore, the standard shapes also define a base anchor, as
well as base west and base east, for placing things on the baseline of the text.
The standard shapes also define a mid anchor (and mid west and mid east). This anchor is half the
height of the character “x” above the base line. This anchor is useful for vertically centering multiple
nodes that have different heights and depth. Here is an example:
x
y
t
x y t
x y t
\begin{tikzpicture}[scale=3,transform shape]
% First, center alignment -> wobbles
\draw[anchor=center] (0,1) node{x} -- (0.5,1) node{y} -- (1,1) node{t};
% Second, base alignment -> no wobble, but too high
\draw[anchor=base] (0,.5) node{x} -- (0.5,.5) node{y} -- (1,.5) node{t};
% Third, mid alignment
\draw[anchor=mid] (0,0) node{x} -- (0.5,0) node{y} -- (1,0) node{t};
\end{tikzpicture}
123
Unfortunately, while perfectly logical, it is often rather counter-intuitive that in order to place a node
above a given point, you need to specify the south anchor. For this reason, there are some useful options
that allow you to select the standard anchors more intuitively:
• above=¸offset) does the same as anchor=south. If the ¸offset) is specified, the node is additionally
shifted upwards by the given ¸offset).
above
\tikz \fill (0,0) circle (2pt) node[above] {above};
above
\tikz \fill (0,0) circle (2pt) node[above=2pt] {above};
• above left=¸offset) does the same as anchor=south east. If the ¸offset) is specified, the node is
additionally shifted upwards and right by ¸offset).
above left
\tikz \fill (0,0) circle (2pt) node[above left] {above left};
above left
\tikz \fill (0,0) circle (2pt) node[above left=2pt] {above left};
• above right=¸offset) does the same as anchor=south west.
above right
\tikz \fill (0,0) circle (2pt) node[above right] {above right};
• left=¸offset) does the same as anchor=east.
left
\tikz \fill (0,0) circle (2pt) node[left] {left};
• right=¸offset) does the same as anchor=west.
• below=¸offset) does the same as anchor=north.
• below left=¸offset) does the same as anchor=north east.
• below right=¸offset) does the same as anchor=north west.
A second set of options behaves similarly, namely the above of, below of, and so on options. They
cause the same anchors to be set as the options without of, however, their parameter is different: You must
provide the name of another node. The current node will then be placed, say, above this specified node at
a distance given by the option node distance.
• above of=¸node) This option causes the node to be placed at the distance node distance above of
¸node). The anchor is center.
a
b
c
d
e
\begin{tikzpicture}[node distance=1cm]
\draw[help lines] (0,0) grid (3,2);
\node (a) {a};
\node (b) [above of=a] {b};
\node (c) [above of=b] {c};
\node (d) [right of=c] {d};
\node (e) [below right of=d] {e};
\end{tikzpicture}
• above left of=¸node) Works like above of, only the node is now put above and left. The
node distance is the Euclidean distance between the two nodes, not the L
1
-distance.
• above right of=¸node) works similarly.
124
• left of=¸node) works similarly.
• right of=¸node) works similarly.
• below of=¸node) works similarly.
• below left of=¸node) works similarly.
• below right of=¸node) works similarly.
• node distance=¸dimension) sets the distance between nodes that are placed using the ... of options.
Note that this distance is the distance between the centers of the nodes, not the distance between their
borders.
13.6 Transformations
It is possible to transform nodes, but, by default, transformations do not apply to nodes. The reason is that
you usually do not want your text to be scaled or rotated even if the main graphic is transformed. Scaling
text is evil, rotating slightly less so.
However, sometimes you do wish to transform a node, for example, it certainly sometimes makes sense
to rotate a node by 90 degrees. There are two ways in which you can achieve this:
1. You can use the following option:
• transform shape causes the current “external” transformation matrix to be applied to the shape.
For example, if you said \tikz[scale=3] and then say node[transform shape] {X}, you will
get a “huge” X in your graphic.
2. You can give transformation option inside the option list of the node. These transformations always
apply to the node.
A
B
A
B
A
B
\begin{tikzpicture}
\tikzstyle{every node}=[draw]
\draw[style=help lines] (0,0) grid (3,2);
\draw (1,0) node{A}
(2,0) node[rotate=90,scale=1.5] {B};
\draw[rotate=30] (1,0) node{A}
(2,0) node[rotate=90,scale=1.5] {B};
\draw[rotate=60] (1,0) node[transform shape] {A}
(2,0) node[transform shape,rotate=90,scale=1.5] {B};
\end{tikzpicture}
13.7 Placing Nodes on a Line or Curve Explicitly
Until now, we always placed node on a coordinate that is mentioned in the path. Often, however, we wish
to place nodes on “the middle” of a line and we do not wish to compute these coordinates “by hand.” To
facilitate such placements, TikZ allows you to specify that a certain node should be somewhere “on” a line.
There are two ways of specifying this: Either explicitly by using the pos option or implicitly by placing the
node “inside” a path operation. These two ways are described in the following.
• pos=¸fraction) When this option is given, the node is not anchored on the last coordinate. Rather, it is
anchored on some point on the line from the previous coordinate to the current point. The ¸fraction)
dictates how “far” on the line the point should be. A ¸fraction) or 0 is the previous coordinate, 1 is
the current one, everything else is in between. In particular, 0.5 is the middle.
Now, what is “the previous line”? This depends on the previous path construction operation.
In the simplest case, the previous path operation was a “line-to” operation, that is, a --¸coordinate)
operation:
0
1/2
9/10
\tikz \draw (0,0) -- (3,1)
node[pos=0]{0} node[pos=0.5]{1/2} node[pos=0.9]{9/10};
125
The next case is the curve-to operation (the .. operation). In this case, the “middle” of the curve,
that is, the position 0.5 is not necessarily the point at the exact half distance on the line. Rather, it
is some point at “time” 0.5 of a point traveling from the start of the curve, where it is at time 0, to
the end of the curve, which it reaches at time 0.5. The “speed” of the point depends on the length
of the support vectors (the vectors that connect the start and end points to the control points). The
exact math is a bit complicated (depending on your point of view, of course); you may wish to consult
a good book on computer graphics and B´ezier curves if you are intrigued.
0
0.125
0.25
0.375
0.5
0.625
0.75
0.875
1
\tikz \draw (0,0) .. controls +(right:3.5cm) and +(right:3.5cm) .. (0,3)
\foreach \p in {0,0.125,...,1} {node[pos=\p]{\p}};
Another interesting case are the horizontal/vertical line-to operations |- and -|. For them, the position
(or time) 0.5 is exactly the corner point.
0
1/2 9/10
\tikz \draw (0,0) |- (3,1)
node[pos=0]{0} node[pos=0.5]{1/2} node[pos=0.9]{9/10};
0 1/2
9/10
\tikz \draw (0,0) -| (3,1)
node[pos=0]{0} node[pos=0.5]{1/2} node[pos=0.9]{9/10};
For all other path construction operations, the position placement does not work, currently. This will
hopefully change in the future (especially for the arc operation).
• auto=¸direction) This option causes an anchor positions to be calculated automatically according to
the following rule. Consider a line between to points. If the ¸direction) is left, then the anchor is
chosen such that the node is to the left of this line. If the ¸direction) is right, then the node is to
the right of this line. Leaving out ¸direction) causes automatic placement to be enabled with the last
value of left or right used. A ¸direction) of false disables automatic placement. This happens
also whenever an anchor is given explicitly by the anchor option or by one of the above, below, etc.
options.
This option only has an effect for nodes that are placed on lines or curves.
a
b
c
d
e
f
g
h
a–b
b–c
c–d
d–e
e–f
f–g
g–h
h–a
\begin{tikzpicture}[scale=.8,auto=left]
\tikzstyle{every node}=[circle,fill=blue!20]
\node (a) at (-1,-2) {a};
\node (b) at ( 1,-2) {b};
\node (c) at ( 2,-1) {c};
\node (d) at ( 2, 1) {d};
\node (e) at ( 1, 2) {e};
\node (f) at (-1, 2) {f};
\node (g) at (-2, 1) {g};
\node (h) at (-2,-1) {h};
\tikzstyle{every node}=[fill=red!20]
\foreach \from/\to in {a/b,b/c,c/d,d/e,e/f,f/g,g/h,h/a}
\draw [->] (\from) -- (\to) node[midway] {\from--\to};
\end{tikzpicture}
• swap This option exchanges the roles of left and right in automatic placement. That is, if left is
the current auto placement, right is set instead and the other way round.
126
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.025
0.2
0.4 0.6
0.8
0.975
\begin{tikzpicture}[auto]
\draw[help lines,use as bounding box] (0,-.5) grid (4,5);
\draw (0.5,0) .. controls (9,6) and (-5,6) .. (3.5,0)
\foreach \pos in {0,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1}
{node [pos=\pos,swap,fill=red!20] {\pos}}
\foreach \pos in {0.025,0.2,0.4,0.6,0.8,0.975}
{node [pos=\pos,fill=blue!20] {\pos}};
\end{tikzpicture}
q
0
q
1
q
2
q
3
0
1
1
0
0
1
\begin{tikzpicture}[shorten >=1pt,node distance=2cm,auto]
\draw[help lines] (0,0) grid (3,2);
\node[state] (q_0) {$q_0$};
\node[state] (q_1) [above right of=q_0] {$q_1$};
\node[state] (q_2) [below right of=q_0] {$q_2$};
\node[state] (q_3) [below right of=q_1] {$q_3$};
\path[->] (q_0) edge node {0} (q_1)
edge node [swap] {1} (q_2)
(q_1) edge node {1} (q_3)
edge [loop above] node {0} ()
(q_2) edge node [swap] {0} (q_3)
edge [loop below] node {1} ();
\end{tikzpicture}
• sloped This option causes the node to be rotated such that a horizontal line becomes a tangent to
the curve. The rotation is normally done in such a way that text is never “upside down.” To get
upside-down text, use can use [rotate=180] or [allow upside down], see below.
0
0
.
2
5
0
.
5
0
.
7
5
1
\tikz \draw (0,0) .. controls +(up:2cm) and +(left:2cm) .. (1,3)
\foreach \p in {0,0.25,...,1} {node[sloped,above,pos=\p]{\p}};
x
y
\begin{tikzpicture}[->]
\draw (0,0) -- (2,0.5) node[midway,sloped,above] {$x$};
\draw (2,-.5) -- (0,0) node[midway,sloped,below] {$y$};
\end{tikzpicture}
• allow upside down=¸true or false) If set to true, TikZ will not “righten” upside down text.
127
0
0
.
2
5
0
.
5
0
.
7
5
1
\tikz [allow upside down]
\draw (0,0) .. controls +(up:2cm) and +(left:2cm) .. (1,3)
\foreach \p in {0,0.25,...,1} {node[sloped,above,pos=\p]{\p}};
x
y
\begin{tikzpicture}[->,allow upside down]
\draw (0,0) -- (2,0.5) node[midway,sloped,above] {$x$};
\draw (2,-.5) -- (0,0) node[midway,sloped,below] {$y$};
\end{tikzpicture}
There exist styles for specifying positions a bit less “technically”:
• style=midway is set to pos=0.5.
at end
very near end
near end
midway
near start
very near start
at start
\tikz \draw (0,0) .. controls +(up:2cm) and +(left:3cm) .. (1,5)
node[at end] {|at end|}
node[very near end] {|very near end|}
node[near end] {|near end|}
node[midway] {|midway|}
node[near start] {|near start|}
node[very near start] {|very near start|}
node[at start] {|at start|};
• style=near start is set to pos=0.25.
• style=near end is set to pos=0.75.
• style=very near start is set to pos=0.125.
• style=very near end is set to pos=0.875.
• style=at start is set to pos=0.
• style=at end is set to pos=1.
13.8 Placing Nodes on a Line or Curve Implicitly
When you wish to place a node on the line (0,0) -- (1,1), it is natural to specify the node not following the
(1,1), but “somewhere in the middle.” This is, indeed, possible and you can write (0,0) -- node{a} (1,1)
to place a node midway between (0,0) and (1,1).
What happens is the following: The syntax of the line-to path operation is actually -- node¸node
specification)¸coordinate). (It is even possible to give multiple nodes in this way.) When the optional
node is encountered, that is, when the -- is directly followed by node, then the specification(s) are read and
“stored away.” Then, after the ¸coordinate) has finally been reached, they are inserted again, but with the
pos option set.
There are two things to note about this: When a node specification is “stored,” its catcodes become
fixed. This means that you cannot use overly complicated verbatim text in them. If you really need, say, a
verbatim text, you will have to put it in a normal node following the coordinate and add the pos option.
Second, which pos is chosen for the node? The position is inherited from the surrounding scope. However,
this holds only for nodes specified in this implicit way. Thus, if you add the option [near end] to a scope,
128
this does not mean that all nodes given in this scope will be put on near the end of lines. Only the nodes
for which an implicit pos is added will be placed near the end. Typically, this is what you want. Here are
some examples that should make this clearer:
A
B
C
D
\begin{tikzpicture}[near end]
\draw (0cm,4em) -- (3cm,4em) node{A};
\draw (0cm,3em) -- node{B} (3cm,3em);
\draw (0cm,2em) -- node[midway] {C} (3cm,2em);
\draw (0cm,1em) -- (3cm,1em) node[midway] {D} ;
\end{tikzpicture}
Like the line-to operation, the curve-to operation .. also allows you to specify nodes “inside” the opera-
tion. After both the first .. and also after the second .. you can place node specifications. Like for the --
operation, these will be collected and then reinserted after the operation with the pos option set.
13.9 The Label and Pin Options
In addition to the node path operation, nodes can also be added using the label and the pin option. This
is mostly useful for simple nodes.
• label=[¸options)]¸angle):¸text) When this option is given to a node operation, it causes another node
to be added to the path after the current node has been finished. This extra node will have the text
¸text). It is placed according to the following rule: Suppose the node currently under construction
is called main node and let us call the label node label node. Then the anchor of label node is
placed at main node.¸angle). The anchor that is chosen depends on the ¸angle). If the ¸angle) lies
between −3

and +3

, then the anchor west is chosen, which causes label node to be placed right of
the right end main node. If ¸angle) lies between 4

and 86

, the anchor south west is chosen, causing
the label node to be placed above and right of the main node; and so on.
my circle
60

−90

\tikz
\node [circle,draw,label=60:$60^\circ$,label=below:$-90^\circ$] {my circle};
As can be seen in the above example, instead of specifying ¸angle) as a number, it is also possible to
use left, right, above, above left, and so on.
You can pass ¸options) to the node label node. For this, you provide the options in square brackets
before the ¸angle). If you do so, you need to add braces around the whole argument of the label
option and this is also the case if you have brackets or commas or semicolons or anything special in
the ¸text).
my circle
X
\tikz \node [circle,draw,label={[red]above:X}] {my circle};
a, b
\begin{tikzpicture}
\node [circle,draw,label={[name=label node]above left:$a,b$}] {};
\draw (label node) -- +(1,1);
\end{tikzpicture}
If you provide multiple label options, then multiple extra label nodes are added in the order they are
given.
The following styles influence how labels are drawn:
129
– label distance=¸distance) The ¸distance) is additionally inserted between the main node and
the label node. The default is 0pt.
my circle X
Y
Z
\tikz[label distance=5mm]
\node [circle,draw,label=right:X,
label=above right:Y,
label=above:Z] {my circle};
– style=every label This style is used in every node created by the label option. The default is
draw=none,fill=none.
• pin=[¸options)]¸angle):¸text) This is option is quite similar to the label option, but there is one
difference: In addition to adding a extra node to the picture, it also adds an edge from this node to
the main node. This causes the node to look like a pin that has been added to the main node:
q
0
\tikz \node [circle,fill=blue!50,minimum size=1cm,pin=60:$q_0$] {};
The meaning of the ¸options) and the ¸angle) and the ¸text) is exactly the same as for the node option.
Only, the options and styles the influence the way pins look are different:
– pin distance=¸distance) This ¸distance) is used instead of the label distance for the distance
between the main node and the label node. The default is 3ex.
my circle X
Y
Z
\tikz[pin distance=1cm]
\node [circle,draw,pin=right:X,
pin=above right:Y,
pin=above:Z] {my circle};
– style=every pin This style is used in every node created by the pin option. The default is
draw=none,fill=none.
– style=every pin edge This style is used in every edge created by the pin optins. The default
is help lines.
my circle X
Y
Z
\tikzstyle{every pin edge}=[<-,shorten <=1pt,snake=snake,line before snake=4pt]
\tikz[pin distance=15mm]
\node [circle,draw,pin=right:X,
pin=above right:Y,
pin=above:Z] {my circle};
– pin edge=¸options) This option can be used to set the options that are to be used in the edge
created by the pin option. The default is empty.
130
my circle X
Z
\tikz[pin distance=10mm]
\node [circle,draw,pin={[pin edge={blue,thick}]right:X},
pin=above:Z] {my circle};
my circle start
\tikzstyle{every pin edge}=[]
\tikzstyle{initial}=[pin={[pin distance=5mm,
pin edge={<-,shorten <=1pt}]left:start}]
\tikz \node [circle,draw,initial] {my circle};
13.10 Connecting Nodes: Using Nodes as Coordinates
Once you have defined a node and given it a name, you can use this name to reference it. This can be
done in two ways, see also Section 10.2.3. Suppose you have said \path(0,0) node(x) {Hello World!};
in order to define a node named x.
1. Once the node x has been defined, you can use (x.¸anchor)) wherever you would normally use a
normal coordinate. This will yield the position at which the given ¸anchor) is in the picture. Note
that transformations do not apply to this coordinate, that is, (x.north) will be the northern anchor
of x even if you have said scale=3 or xshift=4cm. This is usually what you would expect.
2. You can also just use (x) as a coordinate. In most cases, this gives the same coordinate as (x.center).
Indeed, if the shape of x is coordinate, then (x) and (x.center) have exactly the same effect.
However, for most other shapes, some path construction operations like -- try to be “clever” when
this they are asked to draw a line from such a coordinate or to such a coordinate. When you say
(x)--(1,1), the -- path operation will not draw a line from the center of x, but from the border of x
in the direction going towards (1,1). Likewise, (1,1)--(x) will also have the line end on the border
in the direction coming from (1,1).
In addition to --, the curve-to path operation .. and the path operations -| and |- will also handle
nodes without anchors correctly. Here is an example, see also Section 10.2.3:
Hello World!

2
1
xdx
label
lab
el
\begin{tikzpicture}
\path (0,0) node (x) {Hello World!}
(3,1) node[circle,draw](y) {$\int_1^2 x \mathrm d x$};
\draw[->,blue] (x) -- (y);
\draw[->,red] (x) -| node[near start,below] {label} (y);
\draw[->,orange] (x) .. controls +(up:1cm) and +(left:1cm) .. node[above,sloped] {label} (y);
\end{tikzpicture}
13.11 Connecting Nodes: Using the Edge Operation
The edge operation works like a to operation that is added after the main path has been drawn, much like
a node is added after the main path has been drawn. This allows you to have each edge to have a different
appearance. As the node operation, an edge temporarily suspends the construction of the current path and
a new path p is constructed. This new path p will be drawn after the main path has been drawn. Note that
p can be totally different from the main path with respect to its options. Also note that if there are several
to and/or node operations in the main path, each creates its own path(s) and they are drawn in the order
that they are encountered on the path.
131
\path . . . edge[¸options)] ¸nodes) (¸coordinate)) . . . ;
The effect of the edge operation is that after the main path the following path is added to the picture:
\path[every edge,¸options)] (\tikztostart) ¸path);
Here, ¸path) is the to path. Note that, unlike the path added by the to operation, the (\tikztostart)
is added before the ¸path) (which is unnecessary for the to operation, since this coordinate is already
part of the main path).
The \tikztostart is the last coordinate on the path just before the edge operation, just as for the node
or to operations. However, there is one exception to this rule: If the edge operation is directly preceded
by a node operation, then this just-declared node is the start coordinate (and not, as would normally
be the case, the coordinate where this just-declared node is placed – a small, but subtle difference). In
this regard, edge differs from both node and to.
If there are several edge operations in a row, the start coordinate is the same for all of them as their
target coordiantes are not, after all, part of the main path. The start coordinate is, thus, the coordinate
preceding the first edge operation. This is similar to nodes insofar as the edge operation does not
modify the current path at all. In particular, it does not change the last coordinate visited, see the
following example:
a
b
c
d
\begin{tikzpicture}
\node (a) at (0:1) {$a$};
\node (b) at (90:1) {$b$} edge [->] (a);
\node (c) at (180:1) {$c$} edge [->] (a)
edge [<-] (b);
\node (d) at (270:1) {$d$} edge [->] (a)
edge [dotted] (b)
edge [<-] (c);
\end{tikzpicture}
A different way of specifying the above graph using the edge operation is the following:
a
b
c
d
\begin{tikzpicture}
\foreach \name/\angle in {a/0,b/90,c/180,d/270}
\node (\name) at (\angle:1) {$\name$};
\path[->] (b) edge (a)
edge (c)
edge [-,dotted] (d)
(c) edge (a)
edge (d)
(d) edge (a);
\end{tikzpicture}
As can be seen, the path of the edge operation inherits the options from the main path, but you can
locally overrule them.
a
b
c
d
5
m
i
s
s
i
n
g
v
e
r
y
b
a
d
\begin{tikzpicture}
\foreach \name/\angle in {a/0,b/90,c/180,d/270}
\node (\name) at (\angle:1.5) {$\name$};
\path[->] (b) edge node[above right] {$5$} (a)
edge (c)
edge [-,dotted] node[below,sloped] {missing} (d)
(c) edge (a)
edge (d)
(d) edge [red] node[above,sloped] {very}
node[below,sloped] {bad} (a);
\end{tikzpicture}
Instead of every to, the style every edge is installed at the beginning of the main path.
• style=every edge this is is draw by default.
\begin{tikzpicture}
\tikzstyle{every to}=[draw,dashed]
\path (0,0) to (3,2);
\end{tikzpicture}
132
13.12 Referencing Nodes Outside the Current Pictures
13.12.1 Referencing a Node in a Different Picture
It is possible (but not quite trivial) to reference nodes in pictures other than the current one. This means
that you can create a picture and a node therein and, later, you can draw a line from some other position
to this node.
To reference nodes in different pictures, proceed as follows:
1. You need to add the remember picture option to all pictures that contain nodes that you wish to
reference and also to all pictures from which you wish to reference a node in another picture.
2. You need to add the overlay option to paths or to whole pictures that contain references to nodes in
different pictures. (This option switches the computation of the bounding box off.)
3. You need to use a driver that supports picture remembering (currently, this is only pdfT
E
X). With
the pdfT
E
X driver you also need to run T
E
X twice.
(For more details on what is going on behind the scenes, see Section 49.3.2.)
Let us have a look at the effect of these options.
• remember picture=¸true or false) This option tells TikZ that it should attempt to remember the
position of the current picture on the page. This attempt may fail depending on which backend driver
is used. Also, even if remembering works, the position may only be available on a second run of T
E
X.
Provided that remebering works, you may consider saying
\tikzsytle{every picture}+=[remember picture]
to make TikZ remember all pictures. This will add one line in the .aux file for each picture in your
document – which typically is not very much. Then, you do not have to worry about remembered
pictures at all.
• overlay This option is mainly intended for use when nodes in other pictures are referenced, but you
can also use it in other situations. The effect of this option is that everything within the current scope
is not taken into consideration when the bounding box of the current picture is computed.
You need to specify this option on all paths (or at least on all parts of paths) that contain a reference
to a node in another picture. The reason is that, otherwise, TikZ will attempt to make the current
picture large enough to encompass the node in the other picture. However, on a second run of T
E
X this
will create an even bigger picture, leading to larger and larger pictures. Unless you know what you
are doing, I suggest specifying the overlay option with all pictures that contain references to other
pictures.
Let us now have a look at a few examples. These examples work only if this document is processed with
a driver that supports picture remembering.
133
Inside the current text we place two pictures, containing nodes named n1 and n2, using
\tikz[remember picture] \node[circle,fill=red!50] (n1) {};
which yields , and
\tikz[remember picture] \node[fill=blue!50] (n2) {};
yielding the node . To connect these nodes, we create another picture using the overlay option and also
the remember picture option.
\begin{tikzpicture}[remember picture,overlay]
\draw[->,very thick] (n1) -- (n2);
\end{tikzpicture}
Note that the last picture is seemingly empty. What happens is that it has zero size and contains an arrow
that lies well outside its bounds. As a last example, we connect a node in another picture to the first two
nodes. Here, we provide the overlay option only with the line that we do not wish to count as part of the
picture.
Big circle
\begin{tikzpicture}[remember picture]
\node (c) [circle,draw] {Big circle};
\draw [overlay,->,very thick,red,opacity=.5]
(c) to[bend left] (n1) (n1) -| (n2);
\end{tikzpicture}
13.12.2 Referencing the Current Page Node – Absolute Positioning
There is a special node called current page that can be used to access the current page. It is a node
of shape rectangle whose south west anchor is the lower left corner of the page and whose north east
anchor is the upper right corner of the page. While this node is handled in a special way internally, you can
reference it as if it were defined in some remembered picture other than the current one. Thus, by giving the
remembered picture and the overlay options to a picture, you can position nodes absolutely on a page.
The first example places some text in the lower left corner of the current page:
This is an absolutely positioned text in the
lower left corner. No shipout-hackery is
used.
\begin{tikzpicture}[remember picture,overlay]
\node [xshift=1cm,yshift=1cm] at (current page.south west)
[text width=7cm,fill=red!20,rounded corners,above right]
{
This is an absolutely positioned text in the
lower left corner. No shipout-hackery is used.
};
\end{tikzpicture}
The next example adds a circle in the middle of the page.
\begin{tikzpicture}[remember picture,overlay]
\draw [line width=1mm,opacity=.25]
(current page.center) circle (3cm);
\end{tikzpicture}
The final example overlays some text over the page (depending on where this example is found on the
page, the text may also be behind the page).
E
x
a
m
p
l
e
\begin{tikzpicture}[remember picture,overlay]
\node [rotate=60,scale=10,text opacity=0.2]
at (current page.center) {Example};
\end{tikzpicture}
13.13 Predefined Shapes
pgf and TikZ define three shapes, by default:
• rectangle,
• circle, and
• coordinate.
134
By loading library packages, you can define more shapes like ellipses or diamonds; see the library section for
the complete list of shapes.
The exact behaviour of these shapes differs, shapes defined for more special purposes (like a, say, transistor
shape) will have even more custom behaviors. However, there are some options that apply to most shapes:
• inner sep=¸dimension) An additional (invisible) separation space of ¸dimension) will be added inside
the shape, between the text and the shape’s background path. The effect is as if you had added
appropriate horizontal and vertical skips at the beginning and end of the text to make it a bit “larger.”
The default inner sep is the size of a normal space.
tight
loose
default
\begin{tikzpicture}
\draw (0,0) node[inner sep=0pt,draw] {tight}
(0cm,2em) node[inner sep=5pt,draw] {loose}
(0cm,4em) node[fill=examplefill] {default};
\end{tikzpicture}
• inner xsep=¸dimension) Specifies the inner separation in the x-direction, only.
• inner ysep=¸dimension) Specifies the inner separation in the y-direction, only.
• outer sep=¸dimension) This option adds an additional (invisible) separation space of ¸dimension)
outside the background path. The main effect of this option is that all anchors will move a little “to
the outside.”
The default for this option is half the line width. When the default is used and when the background
path is draw, the anchors will lie exactly on the “outside border” of the path (not on the path itself).
When the shape is filled, but not drawn, this may not be desirable. In this case, the outer sep should
be set to zero point.
filled drawn
\begin{tikzpicture}
\draw[line width=5pt]
(0,0) node[outer sep=0pt,fill=examplefill] (f) {filled}
(2,0) node[inner sep=.5\pgflinewidth+2pt,draw] (d) {drawn};
\draw[->] (1,-1) -- (f);
\draw[->] (1,-1) -- (d);
\end{tikzpicture}
• outer xsep=¸dimension) Specifies the outer separation in the x-direction, only.
• outer ysep=¸dimension) Specifies the outer separation in the y-direction, only.
• minimum height=¸dimension) This option ensures that the height of the shape (including the inner,
but ignoring the outer separation) will be at least ¸dimension). Thus, if the text plus the inner
separation is not at least as large as ¸dimension), the shape will be enlarged appropriately. However,
if the text is already larger than ¸dimension), the shape will not be shrunk.
1cm 0cm
\begin{tikzpicture}
\draw (0,0) node[minimum height=1cm,draw] {1cm}
(2,0) node[minimum height=0cm,draw] {0cm};
\end{tikzpicture}
• minimum width=¸dimension) same as minimum height, only for the width.
3 2
\begin{tikzpicture}
\draw (0,0) node[minimum height=2cm,minimum width=3cm,draw] {$3 \times 2$};
\end{tikzpicture}
135
• minimum size=¸dimension) sets both the minimum height and width at the same time.
square
circle
\begin{tikzpicture}
\draw (0,0) node[minimum size=2cm,draw] {square};
\draw (0,-2) node[minimum size=2cm,draw,circle] {circle};
\end{tikzpicture}
• aspect=¸aspect ratio) sets a desired aspect ratio for the shape. For the diamond shape, this option
sets the ratio between width and height of the shape.
aspect 1
aspect 2
\begin{tikzpicture}
\draw (0,0) node[aspect=1,diamond,draw] {aspect 1};
\draw (0,-2) node[aspect=2,diamond,draw] {aspect 2};
\end{tikzpicture}
The coordinate shape is handled in a special way by TikZ. When a node x whose shape is coordinate
is used as a coordinate (x), this has the same effect as if you had said (x.center). None of the special “line
shortening rules” apply in this case. This can be useful since, normally, the line shortening causes paths to
be segmented and they cannot be used for filling. Here is an example that demonstrates the difference:
\begin{tikzpicture}
\tikzstyle{every node}=[draw]
\path[yshift=1.5cm,shape=rectangle]
(0,0) node(a1){} (1,0) node(a2){}
(1,1) node(a3){} (0,1) node(a4){};
\filldraw[fill=examplefill] (a1) -- (a2) -- (a3) -- (a4);
\path[shape=coordinate]
(0,0) coordinate(b1) (1,0) coordinate(b2)
(1,1) coordinate(b3) (0,1) coordinate(b4);
\filldraw[fill=examplefill] (b1) -- (b2) -- (b3) -- (b4);
\end{tikzpicture}
13.14 Executing Code After Nodes
It is possible to add a path right after a node using the option after node path. The idea is that a style
might use this option to add some additional stuff to the node that has just been typeset.
• after node path=¸path) The ¸path) is added to the main path right after the node, as if you had
given the path thereafter. This option can only be given inside the option list of a node and multiple
calls of this option accumulate.
Inside the ¸path) you have access to the node that has just been created via the macro \tikzlastnode.
136
hello
\tikz
\draw node [draw,after node path={(\tikzlastnode) circle (2cm)}]
{hello};
Note that in the above example, if we had written \path instead of \draw, the circle would not have
been drawn since the circle is part of the main path, not part of the node itself.
\tikzaddafternodepathoption{¸code)}
This command allows you to specify that the ¸code) should be executed at the beginning of the
after node path of the current node. The code will also be executed immediately, but also again
at the beginning of an after node path.
137
14 Matrices and Alignment
14.1 Overview
When creating pictures, one often faces the problem of correctly aligning parts of the picture. For example,
you might wish that the base lines of certain nodes should be on the same line and some further nodes should
be below these nodes with, say, their centers on a vertical lines. There are different ways of solving such
problems. For example, by making clever use of anchors, nearly all such alignment problems can be solved.
However, this often leads to complicated code. An often simpler way is to use matrices, the use of which is
explaied in the current section.
A TikZ matrix is similar to L
A
T
E
X’s {tabular} or {array} environment, only instead of text each cell
contains a little picture or a node. The sizes of the cells are automatically adjusted such that they are large
enough to contain all the cell contents.
Matrices are a powerful tool and they need to handled with some care. For impatient readers who skip
the rest of this section: you must end every row with \\. In particular, the last row must be ended with \\.
Many of the ideas implemented in TikZ’s matrix support are due to Mark Wibrow – many thanks to
Mark at this point!
14.2 Matrices are Nodes
Matrices are special in many ways, but for most purposes matrices are treated like nodes. This means, that
you use the node path command to create a matrix and you only use a special option, namely the matrix
option, to signal that the node will contain a matrix. Instead of the usual T
E
X-box that makes up the text
part of the node’s shape, the matrix is used. Thus, in particular, a matrix can have a shape, this shape
can be drawn or filled, it can be used in a tree, and so on. Also, you can refer to the different anchors of a
matrix.
• matrix=¸true or false) This option can be passed to a node path command. It signals that the node
will contain a matrix. The default parameter is true and should usually be omitted.
H
ello
\begin{tikzpicture}
\draw[help lines] (0,0) grid (4,2);
\node [matrix,fill=red!20,draw=blue,very thick] (my matrix) at (2,1)
{
\draw (0,0) circle (4mm); & \node[rotate=10] {Hello}; \\
\draw (0.2,0) circle (2mm); & \fill[red] (0,0) circle (3mm); \\
};
\draw [very thick,->] (0,0) |- (my matrix.west);
\end{tikzpicture}
The exact syntax of the matrix is explained in the course of this section.
• style=every matrix This style is used in every matrix. It is empty by default.
Even more so than nodes, matrices will often be the only object on a path. Because of this, there is a
special abbreviation for creating matrices:
\matrix
Inside {tikzpicture} this is an abbreviation for \path node[matrix].
Even though matrices are nodes, some options do not have the same effect as for normal nodes:
1. Rotations and scaling have no effect on a matrix as a whole (however, you can still transform the
contents of the cells normally). Before the matrix is typeset, the rotational and scaling part of the
transformation matrix is reset.
2. For multi-part shapes you can only set the text part of the node.
3. All options starting with text such as text width have no effect.
138
14.3 Cell Pictures
A matrix consists of rows of cells. Each row (including the last one!) is ended by the command \\. The
character & is used to separate cells. Inside each cell, you must place commands for drawing a picture,
called the cell picture in the following. (However, cell pictures are not enclosed in a complete {pgfpicture}
environment, they are a bit more light-weight. The main difference is that cell pictures cannot have layers.)
It is not necessary to specify beforehand how many rows or columns there are going to be and if a row
contains less cell pictures than another line, empty cells are automatically added as needed.
14.3.1 Alignment of Cell Pictures
For each cell picture a bounding box is computed. These bounding boxes and the origins of the cell pictures
determine how the cells are aligned. Let us start with the rows: Consider the cell pictures on the first row.
Each has a bounding box and somewhere inside this bounding box the origin of the cell picture can be found
(the origin might even lie outside the bounding box, but let us ignore this problem for the moment). The
cell pictures are then shifted around such that all origins lie on the same horizontal line. This may make it
necessary to shift some cell pictures upwards and other downwards, but it can be done and this yields the
vertical alignment of the cell pictures this row. The top of the row is then given by the top of the “highest”
cell picture in the row, the bottom of the row is given by the bottom of the lowest cell picture. (To be more
precise, the height of the row is the maximum y-value of any of the bounding boxes and the depth of the
row is the negated minimum y-value of the bounding boxes).
a X g
\begin{tikzpicture}
\tikzstyle{every node}=[draw=black,anchor=base,font=\huge]
\matrix [draw=red]
{
\node {a}; \fill[blue] (0,0) circle (2pt); &
\node {X}; \fill[blue] (0,0) circle (2pt); &
\node {g}; \fill[blue] (0,0) circle (2pt); \\
};
\end{tikzpicture}
Each row is aligned in this fashion: For each row the cell pictures are vertically aligned such that the
origins lie on the same line. Then the second row is placed below the first row such that the bottom of the
first row touches the top of the second row (unless a row sep is used to add a bit of space). Then the bottom
of the second row touches the top of the third row, and so on. Typically, each row will have an individual
height and depth.
a X g
a X g
a X g
a X g
\begin{tikzpicture}
\tikzstyle{every node}=[draw=black,anchor=base]
\matrix [draw=red]
{
\node {a}; & \node {X}; & \node {g}; \\
\node {a}; & \node {X}; & \node {g}; \\
};
\matrix [row sep=3mm,draw=red] at (0,-2)
{
\node {a}; & \node {X}; & \node {g}; \\
\node {a}; & \node {X}; & \node {g}; \\
};
\end{tikzpicture}
Let us now have a look at the columns. The rules for how the pictures on any given column are aligned are
very similar to the row alignment: Consider all cell pictures in the first column. Each is shifted horizontally
such that the origins lie on the same vertical line. Then, the left end of the column is at the left end of the
bounding box that protrudes furthest to the left. The right end of the column is at the right end of the
bounding box that protrudes furthest to the left. This fixes the horizontal alignment of the cell pictures in
the first column and the same happens the cell pictures in the other columns. Then, the right end of the
first column touches the left end of the second column (unless column sep is used). The right end of the
second column touches the left end of the third column, and so on. (Internally, two columns are actually
used to achieve the desired horizontal alignment, but that is only and implementation detail.)
139
Hallo
X
g
\begin{tikzpicture}
\tikzstyle{every node}=[draw]
\matrix [draw=red]
{
\node[left] {Hallo}; \fill[blue] (0,0) circle (2pt); \\
\node {X}; \fill[blue] (0,0) circle (2pt); \\
\node[right] {g}; \fill[blue] (0,0) circle (2pt); \\
};
\end{tikzpicture}
8 1 6
3 5 7
4 9 2
\begin{tikzpicture}
\tikzstyle{every node}=[draw]
\matrix [draw=red,column sep=1cm]
{
\node {8}; & \node{1}; & \node {6}; \\
\node {3}; & \node{5}; & \node {7}; \\
\node {4}; & \node{9}; & \node {2}; \\
};
\end{tikzpicture}
14.3.2 Setting and Adjusting Column and Row Spacing
There are different ways of setting and adjusting the spacing between columns and rows. First, you can use
the options column sep and row sep to set a default spacing for all rows and all columns. Second, you can
add options to the & character and the \\ command to adjust the spacing between two specific columns or
rows. Additionally, you can specify whether the space between two columns or rows should be considered
between the origins of cells in the column or row or between their borders.
• column sep=¸spacing list) This option sets a default space that is added between every two columns.
This space can be positive or negative and is zero by default. The ¸spacing list) normally contains a
single dimension like 2pt.
123 1 1
12 12 1
1 123 1
1cm
\begin{tikzpicture}
\matrix [draw,column sep=1cm,nodes=draw]
{
\node(a) {123}; & \node (b) {1}; & \node {1}; \\
\node {12}; & \node {12}; & \node {1}; \\
\node(c) {1}; & \node (d) {123}; & \node {1}; \\
};
\draw [red,thick] (a.east) -- (a.east |- c)
(d.west) -- (d.west |- b);
\draw [<->,red,thick] (a.east) -- (d.west |- b)
node [above,midway] {1cm};
\end{tikzpicture}
More generally, the ¸spacing list) may contain a whole list of numbers, separated by commas, and
occurrences of the two key words between origins and between borders. The effect of specifying
such a list is the following: First, all numbers occurring in the list are simply added to compute the
final spacing. Second, concerning the two keywords, the last occurrence of one of the keywords is
important. If the last occurrence is between borders or if neither occurs, then the space is inserted
between the two columns normally. However, if the last occurs is between origins, then the following
happens: The distance between the columns is adjusted such that the difference between the origins
of all the cells in the first column (remember that they all lie on straight line) and the origins of all
the cells in the second column is exactly the given distance.
The between origins option can only be used for columns mentioned in the first row, that is, you
cannot specify this option for columns introduced only in later rows.
140
123 1 1
12 12 1
1 123 1
1cm
\begin{tikzpicture}
\matrix [draw,column sep={1cm,between origins},nodes=draw]
{
\node(a) {123}; & \node (b) {1}; & \node {1}; \\
\node {12}; & \node {12}; & \node {1}; \\
\node {1}; & \node {123}; & \node {1}; \\
};
\draw [<->,red,thick] (a.center) -- (b.center) node [above,midway] {1cm};
\end{tikzpicture}
• row sep=¸spacing list) This option works like column sep, only for rows. Here, too, you can specify
whether the space is added between the lower end of the first row and the upper end of the second
row, or whether the space is computed between the origins of the two rows.
123 1 1
12 12 1
1 123 1
1cm
\begin{tikzpicture}
\matrix [draw,row sep=1cm,nodes=draw]
{
\node (a) {123}; & \node {1}; & \node {1}; \\
\node (b) {12}; & \node {12}; & \node {1}; \\
\node {1}; & \node {123}; & \node {1}; \\
};
\draw [<->,red,thick] (a.south) -- (b.north) node [right,midway] {1cm};
\end{tikzpicture}
123 1 1
12 12 1
1 123 1
1cm
\begin{tikzpicture}
\matrix [draw,row sep={1cm,between origins},nodes=draw]
{
\node (a) {123}; & \node {1}; & \node {1}; \\
\node (b) {12}; & \node {12}; & \node {1}; \\
\node {1}; & \node {123}; & \node {1}; \\
};
\draw [<->,red,thick] (a.center) -- (b.center) node [right,midway] {1cm};
\end{tikzpicture}
The row-end command \\ allows you to provide an optional argument, which must be a dimension.
This dimension will be added to the list in row sep. This means that, firstly, any numbers you list in
this argument will be added as an extra row separation between the line being ended and the next line
and, secondly, you can use the keywords between origins and between borders to locally overrule the
standard setting for this line pair.
1cm
\begin{tikzpicture}
\matrix [row sep=1mm]
{
\draw (0,0) circle (2mm); & \draw (0,0) circle (2mm); \\
\draw (0,0) circle (2mm); & \draw (0,0) circle (2mm); \\[-1mm]
\draw (0,0) coordinate (a) circle (2mm); &
\draw (0,0) circle (2mm); \\[1cm,between origins]
\draw (0,0) coordinate (b) circle (2mm); &
\draw (0,0) circle (2mm); \\
};
\draw [<->,red,thick] (a.center) -- (b.center) node [right,midway] {1cm};
\end{tikzpicture}
The cell separation character & also takes an optional argument, which must also be a spacing list. This
spacing list is added to the column sep having a similar effect as the option for the \\ command for rows.
This optional spacing list can only be given the first time a new column is started (usually in the first
row), subsequent usages of this option in later rows have no effect.
8 1 6
3 5 7
4 9 2
\begin{tikzpicture}
\matrix [draw,nodes=draw,column sep=1mm]
{
\node {8}; &[2mm] \node{1}; &[-1mm] \node {6}; \\
\node {3}; & \node{5}; & \node {7}; \\
\node {4}; & \node{9}; & \node {2}; \\
};
\end{tikzpicture}
141
8 1 6
3 5 7
4 9 2
11mm
\begin{tikzpicture}
\matrix [draw,nodes=draw,column sep=1mm]
{
\node {8}; &[2mm] \node(a){1}; &[1cm,between origins] \node(b){6}; \\
\node {3}; & \node {5}; & \node {7}; \\
\node {4}; & \node {9}; & \node {2}; \\
};
\draw [<->,red,thick] (a.center) -- (b.center) node [above,midway] {11mm};
\end{tikzpicture}
8 1 6
3 5 7
4 9 2
10mm 10mm
\begin{tikzpicture}
\matrix [draw,nodes=draw,column sep={1cm,between origins}]
{
\node (a) {8}; & \node (b) {1}; &[between borders] \node (c) {6}; \\
\node {3}; & \node {5}; & \node {7}; \\
\node {4}; & \node {9}; & \node {2}; \\
};
\draw [<->,red,thick] (a.center) -- (b.center) node [above,midway] {10mm};
\draw [<->,red,thick] (b.east) -- (c.west) node [above,midway] {10mm};
\end{tikzpicture}
14.3.3 Cell Styles and Options
For following style and option are useful for changing the appearance of the all cell pictures:
• style=every cell This style is installed at the beginning of each cell picture. Note that setting this
style to draw will not cause all nodes to be drawn since the draw option has to be passed to each node
individually.
Inside this style (and inside all cells), the current row and column number are accessible via the counters
\pgfmatrixcurrentrow and \pgfmatrixcurrentcolumn.
• cells=¸options) This option adds the ¸options) to the style every cell. It is just a shorthand for
set style={{every cell}+=[¸options)]}.
• nodes=¸options) This option adds the ¸options) to the style every node. It is just a shorthand for
set style={{every node}+=[¸options)]}.
The main use of this option is the install some options for the nodes inside the matrix that should not
apply to the matrix itself.
8 1 6
3 5 7
4 9 2
\begin{tikzpicture}
\matrix [nodes={fill=blue!20,minimum size=5mm}]
{
\node {8}; & \node{1}; & \node {6}; \\
\node {3}; & \node{5}; & \node {7}; \\
\node {4}; & \node{9}; & \node {2}; \\
};
\end{tikzpicture}
The next set of styles can be used to change the appearance of certain rows, columns, or cells. If more
than one of these styles is defined, they are executed in the below order (the every cell style is executed
before all of the below).
• style=column ¸number ) This style is used for every cell in column ¸number).
• style=every odd column This style is used for every cell in an odd column.
• style=every even column This style is used for every cell in an even column.
• style=row ¸number ) This style is used for every cell in row ¸number).
• style=every odd row This style is used for every cell in an odd row.
• style=every even row This style is used for every cell in an even row.
• style=row ¸row number ) column ¸column number ) This style is used for the cell in row ¸row
number) and column ¸column number).
142
8 1 6
3 5 7
4 9 2
\begin{tikzpicture}
\tikzstyle{row 1}=[red]
\tikzstyle{column 2}=[green!50!black]
\tikzstyle{row 3 column 3}=[blue]
\matrix
{
\node {8}; & \node{1}; & \node {6}; \\
\node {3}; & \node{5}; & \node {7}; \\
\node {4}; & \node{9}; & \node {2}; \\
};
\end{tikzpicture}
You can use the column ¸number) option to change the alignment for different columns.
123 456 789
12 45 78
1 4 7
\begin{tikzpicture}
\tikzstyle{column 1}=[anchor=base west]
\tikzstyle{column 2}=[anchor=base east]
\tikzstyle{column 3}=[anchor=base]
\matrix
{
\node {123}; & \node{456}; & \node {789}; \\
\node {12}; & \node{45}; & \node {78}; \\
\node {1}; & \node{4}; & \node {7}; \\
};
\end{tikzpicture}
In many matrices all cell pictures have nearly the same code. For example, cells typically start with
\node{ and end };. The following options allow you to execute such code in all cells:
• execute at begin cell=¸code) The code will be executed at the beginning of each nonempty cell.
• execute at end cell=¸code) The code will be executed at the end of each nonempty cell.
• execute at empty cell=¸code) The code will be executed inside each empty cell.
8 1 6
3 5 7
4 9 2
\begin{tikzpicture}
\tikzstyle{matrix of nodes}=[
execute at begin cell=\node\bgroup,
execute at end cell=\egroup;%
]
\matrix [matrix of nodes]
{
8 & 1 & 6 \\
3 & 5 & 7 \\
4 & 9 & 2 \\
};
\end{tikzpicture}
8 1

3

7
– –
2
\begin{tikzpicture}
\tikzstyle{matrix of nodes}=[
execute at begin cell=\node\bgroup,
execute at end cell=\egroup;,%
execute at empty cell=\node{--};%
]
\matrix [matrix of nodes]
{
8 & 1 & \\
3 & & 7 \\
& & 2 \\
};
\end{tikzpicture}
The matrix library defines a number of styles that make use of the above options.
14.4 Anchoring a Matrix
Since matrices are nodes, they can be anchored in the usual fashion using the anchor option. However, there
are two ways to influence this placement further. First, the following option is often useful:
143
• matrix anchor=¸anchor) This option has the same effect as anchor, but the option applies only to
the matrix itself, not to the cells inside. If you just say anchor=north as an option to the matrix node,
all nodes inside matrix will also have this anchor, unless it is explicitly set differently for each node.
By comparison, matrix anchor sets the anchor for the matrix, but for the nodes inside the value of
anchor remain unchanged.
123
12
1
123
12
1
\begin{tikzpicture}
\matrix [matrix anchor=west] at (0,0)
{
\node {123}; \\ % still center anchor
\node {12}; \\
\node {1}; \\
};
\matrix [anchor=west] at (0,-2)
{
\node {123}; \\ % inherited west anchor
\node {12}; \\
\node {1}; \\
};
\end{tikzpicture}
The second way to anchor a matrix is to use an anchor of a node inside the matrix. For this, the anchor
option has a special effect when given as an argument to a matrix:
• anchor=¸anchor or node.anchor) Normally, the argument of this option refers to anchor of the matrix
node, which is the node than includes all of the stuff of the matrix. However, you can also provide
an argument of the form ¸node).¸anchor) where ¸node) must be node defined inside the matrix and
¸anchor) is an anchor of this node. In this case, the whole matrix is shifted around in such a way that
this particular anchor of this particular node lies at the at position of the matrix. The same is true
for matrix anchor.
a b c d
a b c d
a b c d
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\matrix[matrix anchor=inner node.south,anchor=base,row sep=3mm] at (1,1)
{
\node {a}; & \node {b}; & \node {c}; & \node {d}; \\
\node {a}; & \node(inner node) {b}; & \node {c}; & \node {d}; \\
\node {a}; & \node {b}; & \node {c}; & \node {d}; \\
};
\draw (inner node.south) circle (1pt);
\end{tikzpicture}
14.5 Considerations Concerning Active Characters
Even though TikZ seems to use & to separate cells, pgf actually uses a different command to separate cells,
namely the command \pgfmatrixnextcell and using a normal & character will normally fail. What happens
is that, TikZ makes & an active character and then defines this character to be equal to \pgfmatrixnextcell.
In most situations this will work nicely, but sometimes & cannot be made active; for instance because the
matrix is used in an argument of some macro or the matrix contains nodes that contain normal {tabular}
environments. In this case you can use the following option to avoid having to type \pgfmatrixnextcell
each time:
• ampersand replacement=¸macro name or empty) If a macro name is provided, this macro will be
defined to be equal to \pgfmatrixnextcell inside matrices and & will not be made active. For
instance, you could say ampersand replacement=\& and then use & to separate columns as in the
following example:
H
ello
\tikz
\matrix [ampersand replacement=\&]
{
\draw (0,0) circle (4mm); \& \node[rotate=10] {Hello}; \\
\draw (0.2,0) circle (2mm); \& \fill[red] (0,0) circle (3mm); \\
};
144
14.6 Examples
The following examples are adapted from code by Mark Wibrow. The first two redraw pictures from Timothy
van Zandt’s PSTricks documentation:
U
X
Z
Y X
Y Z
x
p
f
g
q
y
\begin{tikzpicture}
\matrix [matrix of math nodes,row sep=1cm]
{
|(U)| U &[2mm] &[8mm] \\
& |(XZY)| X \times_Z Y & |(X)| X \\
& |(Y)| Y & |(Z)| Z \\
};
\tikzstyle{every node}=[midway,auto,font=\scriptsize]
\draw [double, dashed] (U) -- node {$x$} (X);
\draw (X) -- node {$p$} (X -| XZY.east)
(X) -- node {$f$} (Z)
-- node {$g$} (Y)
-- node {$q$} (XZY)
-- node {$y$} (U);
\end{tikzpicture}
A
B E C
D
g
f
c
b
a
b
e
\begin{tikzpicture}[>=stealth,->,shorten >=2pt,looseness=.5,auto]
\matrix [matrix of math nodes,
column sep={2cm,between origins},
row sep={3cm,between origins},
nodes={circle, draw, minimum size=7.5mm}]
{
& |(A)| A & \\
|(B)| B & |(E)| E & |(C)| C \\
& |(D)| D \\
};
\tikzstyle{every node}=[font=\small\itshape]
\draw (A) to [bend left] (B) node [midway] {g};
\draw (B) to [bend left] (A) node [midway] {f};
\draw (D) -- (B) node [midway] {c};
\draw (E) -- (B) node [midway] {b};
\draw (E) -- (C) node [near end] {a};
\draw [-,line width=8pt,draw=graphicbackground]
(D) to [bend right, looseness=1] (A);
\draw (D) to [bend right, looseness=1] (A) node [near start] {b} node [near end] {e};
\end{tikzpicture}
145
\begin{tikzpicture}
\matrix (network)
[matrix of nodes,%
nodes in empty cells,
nodes={outer sep=0pt,circle,minimum size=4pt,draw},
column sep={1cm,between origins},
row sep={1cm,between origins}]
{
& & & \\
& & & \\
|[draw=none]| & |[xshift=1mm]| & |[xshift=-1mm]| \\
};
\foreach \a in {1,...,4}{
\draw (network-3-2) -- (network-2-\a);
\draw (network-3-3) -- (network-2-\a);
\draw [-stealth] ([yshift=5mm]network-1-\a.north) -- (network-1-\a);
\foreach \b in {1,...,4}
\draw (network-1-\a) -- (network-2-\b);
}
\draw [stealth-] ([yshift=-5mm]network-3-2.south) -- (network-3-2);
\draw [stealth-] ([yshift=-5mm]network-3-3.south) -- (network-3-3);
\end{tikzpicture}
The following example is adapted from code written by Kjell Magne Fauske, which is based on the fol-
lowing paper: K. Bossley, M. Brown, and C. Harris, Neurofuzzy identification of an autonomous underwater
vehicle, International Journal of Systems Science, 1999, 30, 901–913.
expert
initialize
model
system
identify
candidate
model
update
model
evaluate
candidate
models
is best
candidate
stop
yes
no
146
\begin{tikzpicture}[auto]
\tikzstyle{decision} = [diamond, draw=blue, thick, fill=blue!20,
text width=4.5em, text badly centered, inner sep=1pt]
\tikzstyle{block} = [rectangle, draw=blue, thick, fill=blue!20,
text width=5em, text centered, rounded corners, minimum height=4em]
\tikzstyle{line} = [draw, thick, -latex’,shorten >=2pt];
\tikzstyle{cloud} = [draw=red, thick, ellipse,fill=red!20, minimum height=2em];
\matrix [column sep=5mm,row sep=7mm]
{
% row 1
\node [cloud] (expert) {expert}; &
\node [block] (init) {initialize model}; &
\node [cloud] (system) {system}; \\
% row 2
& \node [block] (identify) {identify candidate model}; & \\
% row 3
\node [block] (update) {update model}; &
\node [block] (evaluate) {evaluate candidate models}; & \\
% row 4
& \node [decision] (decide) {is best candidate}; & \\
% row 5
& \node [block] (stop) {stop}; & \\
};
\tikzstyle{every path}=[line]
\path (init) -- (identify);
\path (identify) -- (evaluate);
\path (evaluate) -- (decide);
\path (update) |- (identify);
\path (decide) -| node [near start] {yes} (update);
\path (decide) -- node [midway] {no} (stop);
\path [dashed] (expert) -- (init);
\path [dashed] (system) -- (init);
\path [dashed] (system) |- (evaluate);
\end{tikzpicture}
147
15 Making Trees Grow
15.1 Introduction to the Child Operation
Trees are a common way of visualizing hierarchical structures. A simple tree looks like this:
root
left right
child child
\begin{tikzpicture}
\node {root}
child {node {left}}
child {node {right}
child {node {child}}
child {node {child}}
};
\end{tikzpicture}
Admittedly, in reality trees are more likely to grow upward and not downward as above. You can tell
whether the author of a paper is a mathematician or a computer scientist by looking at the direction their
trees grow. A computer scientist’s trees will grow downward while a mathematician’s tree will grow upward.
Naturally, the correct way is the mathematician’s way, which can be specify as follows:
root
left right
child child
\begin{tikzpicture}
\node {root} [grow’=up]
child {node {left}}
child {node {right}
child {node {child}}
child {node {child}}
};
\end{tikzpicture}
In TikZ, trees are specified by adding children to a node on a path using the child operation:
\path . . . child[¸options)]foreach¸variables)in{¸values)}{¸child path)} . . . ;
This operation should directly follow a completed node operation or another child operation, although
it is permissible that the first child operation is preceded by options (we will come to that).
When a node operation like node {X} is followed by child, TikZ starts counting the number of child
nodes that follow the original node {X}. For this, it scans the input and stores away each child and
its arguments until it reaches a path operation that is not a child. Note that this will fix the character
codes of all text inside the child arguments, which means, in essence, that you cannot use verbatim text
inside the nodes inside a child. Sorry.
Once the children have been collected and counted, TikZ starts generating the child nodes. For each
child of a parent node TikZ computes an appropriate position where the child is placed. For each child,
the coordinate system is transformed so that the origin is at this position. Then the ¸child path) is
drawn. Typically, the child path just consists of a node specification, which results in a node being
drawn at the child’s position. Finally, an edge is drawn from the first node in the ¸child path) to the
parent node.
The optional foreach part (note that there is no backslash before foreach) allows you to specify
multiple children in a single child command. The idea is the following: A \foreach statement is
(internally) used to iterate over the list of ¸values). For each value in this list, a new child is added to
the node. The syntax for ¸variables) and for ¸values) is the same as for the \foreach statement, see
Section 34. For example, when you say
node {root} child [red] foreach \name in {1,2} {node {\name}}
the effect will be the same as if you had said
node {root} child[red] {node {1}} child[ref] {node {2}}
When you write
node {root} child[\pos] foreach \name/\pos in {1/left,2/right} {node[\pos] {\name}}
148
the effect will be the same as for
node {root} child[left] {node[left] {1}} child[right] {node[right] {2}}
You can nest things as in the following example:
\begin{tikzpicture}[level distance=4mm]
\tikzstyle{level 1}=[sibling distance=8mm]
\tikzstyle{level 2}=[sibling distance=4mm]
\tikzstyle{level 3}=[sibling distance=2mm]
\coordinate
child foreach \x in {0,1}
{child foreach \y in {0,1}
{child foreach \z in {0,1}}};
\end{tikzpicture}
The details and options for this operation are described in the rest of this present section.
15.2 Child Paths and the Child Nodes
For each child of a root node, its ¸child path) is inserted at a specific location in the picture (the placement
rules are discussed in Section 15.5). The first node in the ¸child path), if it exists, is special and called
the child node. If there is no first node in the ¸child path), that is, if the ¸child path) is missing (including
the curly braces) or if it does not start with node or with coordinate, then an empty child node of shape
coordinate is automatically added.
Consider the example \node {x} child {node {y}} child;. For the first child, the ¸child path) has
the child node node {y}. For the second child, no child node is specified and, thus, it is just coordinate.
As for any normal node, you can give the child node a name, shift it around, or use options to influence
how it is rendered.
root
left right
\begin{tikzpicture}
\node[rectangle,draw] {root}
child {node[circle,draw] (left node) {left}}
child {node[ellipse,draw] (right node) {right}};
\draw[dashed,->] (left node) -- (right node);
\end{tikzpicture}
In many cases, the ¸child path) will just consist of a specification of a child node and, possibly, children
of this child node. However, the node specification may be followed by arbitrary other material that will be
added to the picture, transformed to the child’s coordinate system. For your convenience, a move-to (0,0)
operation is inserted automatically at the beginning of the path. Here is an example:
root
\begin{tikzpicture}
\node {root}
child {[fill] circle (2pt)}
child {[fill] circle (2pt)};
\end{tikzpicture}
At the end of the ¸child path) you may add a special path operation called edge from parent. If this
operation is not given by yourself somewhere on the path, it will be automatically added at the end. This
option causes a connecting edge from the parent node to the child node to be added to the path. By
giving options to this operation you can influence how the edge is rendered. Also, nodes following the
edge from parent operation will be placed on this edge, see Section 15.6 for details.
To sum up:
1. The child path starts with a node specification. If it is not there, it is added automatically.
2. The child path ends with a edge from parent operation, possibly followed by nodes to be put on this
edge. If the operation is not given at the end, it is added automatically.
15.3 Naming Child Nodes
Child nodes can be named like any other node using either the name option or the special syntax in which
the name of the node is placed in round parentheses between the node operation and the node’s text.
149
If you do not assign a name to a child node, TikZ will automatically assign a name as follows: Assume
that the name of the parent node is, say, parent. (If you did not assign a name to the parent, TikZ will do
so itself, but that name will not be user-accessible.) The first child of parent will be named parent-1, the
second child is named parent-2, and so on.
This naming convention works recursively. If the second child parent-2 has children, then the first of
these children will be called parent-2-1 and the second parent-2-2 and so on.
If you assign a name to a child node yourself, no name is generated automatically (the node does not have
two names). However, “counting continues,” which means that the third child of parent is called parent-3
independently of whether you have assigned names to the first and/or second child of parent.
Here is an example:
root
root-1 root-2
special root-2-2
\begin{tikzpicture}
\node (root) {root}
child
child {
child {coordinate (special)}
child
};
\node at (root-1) {root-1};
\node at (root-2) {root-2};
\node at (special) {special};
\node at (root-2-2) {root-2-2};
\end{tikzpicture}
15.4 Specifying Options for Trees and Children
Each child may have its own ¸options), which apply to “the whole child,” including all of its grandchildren.
Here is an example:
\begin{tikzpicture}[thick]
\tikzstyle{level 2}=[sibling distance=10mm]
\coordinate
child[red] {child child}
child[green] {child child[blue]};
\end{tikzpicture}
The options of the root node have no effect on the children since the options of a node are always “local”
to that node. Because of this, the edges in the following tree are black, not red.
root
\begin{tikzpicture}[thick]
\node [red] {root}
child
child;
\end{tikzpicture}
This raises the problem of how to set options for all children. Naturally, you could always set options for
the whole path as in \path [red] node {root} child child; but this is bothersome in some situations.
Instead, it is easier to give the options before the first child as follows:
root
\begin{tikzpicture}[thick]
\node [red] {root}
[green] % option applies to all children
child
child;
\end{tikzpicture}
Here is the set of rules:
1. Options for the whole tree are given before the root node.
2. Options for the root node are given directly to the node operation of the root.
3. Options for all children can be given between the root node and the first child.
4. Options applying to a specific child path are given as options to the child operation.
150
5. Options applying to the node of a child, but not to the whole child path, are given as options to the
node command inside the ¸child path).
\begin{tikzpicture}
\path
[...] % Options apply to the whole tree
node[...] {root} % Options apply to the root node only
[...] % Options apply to all children
child[...] % Options apply to this child and all its children
{
node[...] {} % Options apply to the child node only
...
}
child[...] % Options apply to this child and all its children
;
\end{tikzpicture}
There are additional styles that influence how children are rendered:
• style=every child This style is used at the beginning of each child, as if you had given the options
to the child operation.
• style=every child node This style is used at the beginning of each child node in addition to the
every node style.
• style=level ¸number ) This style is used at the beginning of each set of children, where ¸number) is
the current level in the current tree. For example, when you say \node {x} child child;, then the
style level 1 is used before the first child. If this first child has children itself, then level 2 would
be used for them.
root
\begin{tikzpicture}
\tikzstyle{level 1}=[sibling distance=20mm]
\tikzstyle{level 2}=[sibling distance=5mm]
\node {root}
child { child child }
child { child child child };
\end{tikzpicture}
15.5 Placing Child Nodes
Perhaps the most difficult part in drawing a tree is the correct layout of the children. Typically, the children
have different sizes and it is not easy to arrange them in such a manner that not too much space is wasted, the
children do not overlap, and they are either evenly spaced or their centers are evenly distributed. Calculating
good positions is especially difficult since a good position for the first child may depend on the size of the
last child.
In TikZ, a comparatively simple approach is taken to placing the children. In order to compute a child’s
position, all that is taken into account is the number of the current child in the list of children and the
number of children in this list. Thus, if a node has five children, then there is a fixed position for the first
child, a position for the second child, and so on. These positions do not depend on the size of the children
and, hence, children can easily overlap. However, since you can use options to shift individual children a bit,
this is not as great a problem as it may seem.
Although the placement of the children only depends on their number in the list of children and the
total number of children, everything else about the placement is highly configurable. You can change the
distance between children (appropriately called the sibling distance) and the distance between levels of
the tree. These distances may change from level to level. The direction in which the tree grows can be
changed globally and for parts of the tree. You can even specify your own “growth function” to arrange
children on a circle or along special lines or curves.
The default growth function works as follows: Assume that we are given a node and five children. These
children will be placed on a line with their centers (or, more generally, with their anchors) spaced apart by
the current sibling distance. The line is orthogonal to the current direction of growth, which is set with
151
the grow and grow’ option (the latter option reverses the ordering of the children). The distance from the
line to the parent node is given by the level distance.
root
1
2
3
4
s
i
b
l
i
n
g
d
i
s
t
a
n
c
e
level
distance
\begin{tikzpicture}
\path [help lines]
node (root) {root}
[grow=-10]
child {node {1}}
child {node {2}}
child {node {3}}
child {node {4}};
\draw[|<->|,thick] (root-1.center)
-- node[above,sloped] {sibling distance} (root-2.center);
\draw[|<->|,thick] (root.center)
-- node[above,sloped] {level distance} +(-10:\tikzleveldistance);
\end{tikzpicture}
Here is a detailed description of the options:
• level distance=¸distance) This option allows you to change the distance between different levels of
the tree, more precisely, between the parent and the line on which its children are arranged. When
given to a single child, this will set the distance for this child only.
root
\begin{tikzpicture}
\node {root}
[level distance=20mm]
child
child {
[level distance=5mm]
child
child
child
}
child[level distance=10mm];
\end{tikzpicture}
root
\begin{tikzpicture}
\tikzstyle{level 1}=[level distance=10mm]
\tikzstyle{level 2}=[level distance=5mm]
\node {root}
child
child {
child
child[level distance=10mm]
child
}
child;
\end{tikzpicture}
• sibling distance=¸distance) This option specifies the distance between the anchors of the children
of a parent node.
\begin{tikzpicture}[level distance=4mm]
\tikzstyle{level 1}=[sibling distance=8mm]
\tikzstyle{level 2}=[sibling distance=4mm]
\tikzstyle{level 3}=[sibling distance=2mm]
\coordinate
child {
child {child child}
child {child child}
}
child {
child {child child}
child {child child}
};
\end{tikzpicture}
152
31
30
20
5 4
10
9 1
20
19
1
18
\begin{tikzpicture}[level distance=10mm]
\tikzstyle{every node}=[fill=red!60,circle,inner sep=1pt]
\tikzstyle{level 1}=[sibling distance=20mm,
set style={{every node}+=[fill=red!45]}]
\tikzstyle{level 2}=[sibling distance=10mm,
set style={{every node}+=[fill=red!30]}]
\tikzstyle{level 3}=[sibling distance=5mm,
set style={{every node}+=[fill=red!15]}]
\node {31}
child {node {30}
child {node {20}
child {node {5}}
child {node {4}}
}
child {node {10}
child {node {9}}
child {node {1}}
}
}
child {node {20}
child {node {19}
child {node {1}}
child[fill=none] {edge from parent[draw=none]}
}
child {node {18}}
};
\end{tikzpicture}
• grow=¸direction) This option is used to define the ¸direction) in which the tree will grow. The
¸direction) can either be an angle in degrees or one of the following special text strings: down, up,
left, right, north, south, east, west, north east, north west, south east, and south west. All
of these have “their obvious meaning,” so, say, south west is the same as the angle −135

.
As a side effect, this option installs the default growth function.
In addition to setting the direction, this option also has a seemingly strange effect: It sets the sibling
distance for the current level to 0pt, but leaves the sibling distance for later levels unchanged.
This somewhat strange behaviour has a highly desirable effect: If you give this option before the list
of children of a node starts, the “current level” is still the parent level. Each child will be on a later
level and, hence, the sibling distance will be as specified originally. This will cause the children to be
neatly aligned in a line orthogonal to the given ¸direction). However, if you give this option locally
to a single child, then “current level” will be the same as the child’s level. The zero sibling distance
will then cause the child to be placed exactly at a point at distance level distance in the direction
¸direction). However, the children of the child will be placed “normally” on a line orthogonal to the
¸direction).
These placement effects are best demonstrated by some examples:
root
\tikz \node {root} [grow=right] child child;
root
\tikz \node {root} [grow=south west] child child;
153
root
\begin{tikzpicture}[level distance=10mm,sibling distance=5mm]
\node {root}
[grow=down]
child
child
child[grow=right] {
child child child
};
\end{tikzpicture}
C
H
H
H
C
H
H
H
This is wrong!
\begin{tikzpicture}[level distance=2em]
\node {C}
child[grow=up] {node {H}}
child[grow=left] {node {H}}
child[grow=down] {node {H}}
child[grow=right] {node {C}
child[grow=up] {node {H}}
child[grow=right] {node {H}}
child[grow=down] {node {H}}
edge from parent[double]
coordinate (wrong)
};
\draw[<-,red] ([yshift=-2mm]wrong) -- +(0,-1)
node[below]{This is wrong!};
\end{tikzpicture}
start node
end
the middle is here
\begin{tikzpicture}
\node[rectangle,draw] (a) at (0,0) {start node};
\node[rectangle,draw] (b) at (2,1) {end};
\draw (a) -- (b)
node[coordinate,midway] {}
child[grow=100,<-] {node[above] {the middle is here}};
\end{tikzpicture}
• grow’=¸direction) This option has the same effect as grow, only the children are arranged in the
opposite order.
• growth parent anchor=¸anchor) This option allows you to specify which anchor of the parent node
is to be used for computing the children’s position. For example, when there is only one child and the
level distance is 2cm, then the child node will be placed two centimeters below the ¸anchor) of the
parent node. “Beinng placed” means that the child node’s anchor (which is the anchor specified using
the anchor= option in the node command of the child) is two centimeters below the parent node’s
¸anchor). The default value of ¸anchor) is center.
In the following example, the two red lines both have length 1cm.
root root
\begin{tikzpicture}[level distance=1cm]
\node [rectangle,draw] (a) at (0,0) {root}
[growth parent anchor=south] child;
\node [rectangle,draw] (b) at (2,0) {root}
[growth parent anchor=north east] child;
\draw [red,thick,dashed] (a.south) -- (a-1);
\draw [red,thick,dashed] (b.north east) -- (b-1);
\end{tikzpicture}
In the next example, the top and bottom nodes are aligned at the top and the bottom.
root
small
big root
big
\begin{tikzpicture}[level distance=2cm,growth parent anchor=north]
\tikzstyle{every node}=[anchor=north,rectangle,draw]
\tikzstyle{every child node}=[anchor=south]
\node at (0,0) {root} child {node {small}};
\node at (2,0) {big root} child {node {\large big}};
\end{tikzpicture}
154
• growth function=¸macro name) This rather low-level option allows you to set a new growth function.
The ¸macro name) must be the name of a macro without parameters. This macro will be called for
each child of a node.
The effect of executing the macro should be the following: It should transform the coordinate system
in such a way that the origin becomes the place where the current child should be anchored. When the
macro is called, the current coordinate system will be setup such that the anchor of the parent node
is in the origin. Thus, in each call, the ¸macro name) must essentially do a shift to the child’s origin.
When the macro is called, the T
E
X counter \tikznumberofchildren will be set to the total number
of children of the parent node and the counter \tikznumberofcurrentchild will be set to the number
of the current child.
The macro may, in addition to shifting the coordinate system, also transform the coordinate system
further. For example, it could be rotated or scaled.
Additional growth functions are defined in the library, see Section 33.
15.6 Edges From the Parent Node
Every child node is connected to its parent node via a special kind of edge called the edge from parent.
This edge is added to the ¸child path) when the following path operation is encountered:
\path . . . edge from parent[¸options)] . . . ;
This path operation can only be used inside ¸child paths) and should be given at the end, possibly
followed by node specifications (we will come to that). If a ¸child path) does not contain this operation,
it will be added at the end of the ¸child path) automatically.
This operation has several effects. The most important is that it inserts the current “edge from parent
path” into the child path. The edge from parent path can be set using the following option:
• edge from parent path=¸path) This options allows you to set the edge from parent path to a new
path. The default for this path is the following:
(\tikzparentnode\tikzparentanchor) -- (\tikzchildnode\tikzchildanchor)
The \tikzparentnode is a macro that will expand to the name of the parent node. This works
even when you have not assigned a name to the parent node, in this case an internal name is
automatically generated. The \tikzchildnode is a macro that expands to the name of the child
node. The two ...anchor macros are empty by default. So, what is essentially inserted is just
the path segment (\tikzparentnode) -- (\tikzchildnode); which is exactly an edge from the
parent to the child.
You can modify this edge from parent path to achieve all sorts of effects. For example, we could
replace the straight line by a curve as follows:
root
left right
child child
\begin{tikzpicture}[edge from parent path=
{(\tikzparentnode.south) .. controls +(0,-1) and +(0,1)
.. (\tikzchildnode.north)}]
\node {root}
child {node {left}}
child {node {right}
child {node {child}}
child {node {child}}
};
\end{tikzpicture}
Further useful edge from parent paths are defined in the tree library, see Section 33.
As said before, the anchors in the default edge from parent path are empty. However, you can set
them using the following options:
– child anchor=¸anchor) Specifies the anchor where the edge from parent meets the child node
by setting the macro \tikzchildanchor to .¸anchor).
If you specify border as the ¸anchor), then the macro \tikzchildanchor is set to the empty
string. The effect of this is that the edge from the parent will meet the child on the border at
an automatically calculated position.
155
root
left right
child child
\begin{tikzpicture}
\node {root}
[child anchor=north]
child {node {left} edge from parent[dashed]}
child {node {right}
child {node {child}}
child {node {child} edge from parent[draw=none]}
};
\end{tikzpicture}
– parent anchor=¸anchor) This option works the same way as the child anchor, only for the
parent.
Besides inserting the edge from parent path, the edge from parent operation has another effect: The
¸options) are inserted directly before the edge from parent path and the following style is also installed
prior to inserting the path:
• style=edge from parent This style is inserted right before the edge from parent path and before
the ¸options) are inserted. By default, it just draws the edge from parent, but you can use it to
make the edge look different.
root
left right
child child
\begin{tikzpicture}
\tikzstyle{edge from parent}=[draw,red,thick]
\node {root}
child {node {left} edge from parent[dashed]}
child {node {right}
child {node {child}}
child {node {child} edge from parent[draw=none]}
};
\end{tikzpicture}
Note: The ¸options) inserted before the edge from parent path is added apply to the whole child path.
Thus, it is not possible to, say, draw a circle in red as part of the child path and then have an edge to
parent in blue. However, as always, the child node is a node and can be drawn in a totally different
way.
Finally, the edge from parent operation has one more effect: It causes all nodes following the operation
to be placed on the edge. This is the same effect as if you had added the pos option to all these nodes,
see also Section 13.7.
As an example, consider the following code:
\node (root) {} child {node (child) {} edge to parent node {label}};
The edge to parent operation and the following node operation will, together, have the same effect
as if we had said:
(root) -- (child) node [pos=0.5] {label}
Here is a more complicated example:
156
root
left
a
b
right
child
c
child
x
\begin{tikzpicture}
\node {root}
child {
node {left}
edge from parent
node[left] {a}
node[right] {b}
}
child {
node {right}
child {
node {child}
edge from parent
node[left] {c}
}
child {node {child}}
edge from parent
node[near end] {x}
};
\end{tikzpicture}
157
16 Plots of Functions
16.1 When Should One Use TikZ for Generating Plots?
There exist many powerful programs that produce plots, examples are gnuplot or mathematica. These
programs can produce two different kinds of output: First, they can output a complete plot picture in
a certain format (like pdf) that includes all low-level commands necessary for drawing the complete plot
(including axes and labels). Second, they can usually also produce “just plain data” in the form of a long
list of coordinates. Most of the powerful programs consider it a to be “a bit boring” to just output tabled
data and very much prefer to produce fancy pictures. Nevertheless, when coaxed, they can also provide the
plain data.
Note that is often not necessary to use Ti kZ for plots. Programs like gnuplot can produce very so-
phisticated plots and it is usually much easier to simply include these plots as a finished pdf or PostScript
graphics.
However, there are a number of reasons why you may wish to invest time and energy into mastering the
pgf commands for creating plots:
• Virtually all plots produced by “external programs” use different fonts from the one used in your
document.
• Even worse, formulas will look totally different, if they can be rendered at all.
• Line width will usually be too large or too small.
• Scaling effects upon inclusion can create a mismatch between sizes in the plot and sizes in the text.
• The automatic grid generated by most programs is mostly distracting.
• The automatic ticks generated by most programs are cryptic numerics. (Try adding a tick reading “π”
at the right point.)
• Most programs make it very easy to create “chart junk” in a most convenient fashion. All show, no
content.
• Arrows and plot marks will almost never match the arrows used in the rest of the document.
The above list is not exhaustive, unfortunately.
16.2 The Plot Path Operation
The plot path operation can be used to append a line or curve to the path that goes through a large number
of coordinates. These coordinates are either given in a simple list of coordinates, read from some file, or
they are computed on the fly.
The syntax of the plot comes in different versions.
\path . . . --plot¸further arguments) . . . ;
This operation plots the curve through the coordinates specified in the ¸further arguments). The current
(sub)path is simply continued, that is, a line-to operation to the first point of the curve is implicitly
added. The details of the ¸further arguments) will be explained in a moment.
\path . . . plot¸further arguments) . . . ;
This operation plots the curve through the coordinates specified in the ¸further arguments) by first
“moving” to the first coordinate of the curve.
The ¸further arguments) are used in three different ways to specifying the coordinates of the points to
be plotted:
1. --plot[¸local options)]coordinates{¸coordinate 1)¸coordinate 2). . . ¸coordinate n)}
2. --plot[¸local options)]file{¸filename)}
3. --plot[¸local options)]¸coordinate expression)
4. --plot[¸local options)]function{¸gnuplot formula)}
These different ways are explained in the following.
158
16.3 Plotting Points Given Inline
In the first two cases, the points are given directly in the T
E
X-file as in the following example:
\tikz \draw plot coordinates {(0,0) (1,1) (2,0) (3,1) (2,1) (10:2cm)};
Here is an example showing the difference between plot and --plot:
\begin{tikzpicture}
\draw (0,0) -- (1,1) plot coordinates {(2,0) (4,0)};
\draw[color=red,xshift=5cm]
(0,0) -- (1,1) -- plot coordinates {(2,0) (4,0)};
\end{tikzpicture}
16.4 Plotting Points Read From an External File
The second way of specifying points is to put them in an external file named ¸filename). Currently, the only
file format that TikZ allows is the following: Each line of the ¸filename) should contain one line starting
with two numbers, separated by a space. Anything following the two numbers on the line is ignored. Also,
lines starting with a % or a # are ignored as well as empty lines. (This is exactly the format that gnuplot
produces when you say set terminal table.) If necessary, more formats will be supported in the future,
but it is usually easy to produce a file containing data in this form.
\tikz \draw plot[mark=x,smooth] file {plots/pgfmanual-sine.table};
The file plots/pgfmanual-sine.table reads:
#Curve 0, 20 points
#x y type
0.00000 0.00000 i
0.52632 0.50235 i
1.05263 0.86873 i
1.57895 0.99997 i
...
9.47368 -0.04889 i
10.00000 -0.54402 i
It was produced from the following source, using gnuplot:
set terminal table
set output "../plots/pgfmanual-sine.table"
set format "%.5f"
set samples 20
plot [x=0:10] sin(x)
The ¸local options) of the plot operation are local to each plot and do not affect other plots “on the same
path.” For example, plot[yshift=1cm] will locally shift the plot 1cm upward. Remember, however, that
most options can only be applied to paths as a whole. For example, plot[red] does not have the effect of
making the plot red. After all, you are trying to “locally” make part of the path red, which is not possible.
16.5 Plotting a Function
When you plot a function, the coordinates of the plot data can be computed by evaluating a mathematical
expression. Since pgf comes with a mathematical engine, you can specify this expression and then have
TikZ produce the desired coordinates for you, automatically.
159
Since this case is quite common when plotting a function, the syntax is easy: Following the plot command
and its local options, you directly provide a ¸coordinate expression). It looks like a normal coordinate, but
inside you may use a special macro, which is \x by default, but this can be changed using the variable
option. The ¸coordinate expression) is then evaluated for different values for \x and the resulting coordinates
are plotted.
Note that you will often have to put the x- or y-coordinate inside braces, namely whenever you use an
expression involving a paranthesis.
The following options influence how the ¸coordinate expression) is evaluated:
• variable=¸macro) sets the macro whose value is set to the different values when ¸coordinate expression)
is evaluated.
• samples=¸number) sets the number of samples used in the plot. The default is 25.
• domain=¸start):¸end) sets the domain between which the samples are taken. The default is -5:5.
• samples at=¸sample list) This option specifies a list of positions for which the variable should be
evaluated. For instance, you can say samples at={1,2,8,9,10} to have the variable evaluated exactly
for values 1, 2, 8, 9, and 10. You can use the \foreach syntax, so you can use ... inside the ¸sample
list).
When this option is used, the samples and domain option are overruled. The other ways round, setting
either samples or domain will overrule this option.
x
f(x)
f(x) = x
f(x) = sin x
f(x) =
1
20
e
x
\begin{tikzpicture}[domain=0:4]
\draw[very thin,color=gray] (-0.1,-1.1) grid (3.9,3.9);
\draw[->] (-0.2,0) -- (4.2,0) node[right] {$x$};
\draw[->] (0,-1.2) -- (0,4.2) node[above] {$f(x)$};
\draw[color=red] plot (\x,\x) node[right] {$f(x) =x$};
\draw[color=blue] plot (\x,{sin(\x r)}) node[right] {$f(x) = \sin x$};
\draw[color=orange] plot (\x,{0.05*exp(\x)}) node[right] {$f(x) = \frac{1}{20} \mathrm e^x$};
\end{tikzpicture}
\tikz \draw[scale=0.5,domain=-3.141:3.141,smooth,variable=\t]
plot ({\t*sin(\t r)},{\t*cos(\t r)});
\tikz \draw[domain=0:360,smooth,variable=\t]
plot ({sin(\t)},\t/360,{cos(\t)});
160
16.6 Plotting a Function Using Gnuplot
Often, you will want to plot points that are given via a function like f(x) = xsin x. Unfortunately, T
E
X
does not really have enough computational power to generate the points on such a function efficiently (it is
a text processing program, after all). However, if you allow it, T
E
X can try to call external programs that
can easily produce the necessary points. Currently, TikZ knows how to call gnuplot.
When TikZ encounters your operation plot[id=¸id)] function{x*sin(x)} for the first time, it will
create a file called ¸prefix)¸id).gnuplot, where ¸prefix) is \jobname. by default, that is, the name of you
main .tex file. If no ¸id) is given, it will be empty, which is alright, but it is better when each plot has
a unique ¸id) for reasons explained in a moment. Next, TikZ writes some initialization code into this file
followed by plot x*sin(x). The initialization code sets up things such that the plot operation will write
the coordinates into another file called ¸prefix)¸id).table. Finally, this table file is read as if you had said
plot file{¸prefix)¸id).table}.
For the plotting mechanism to work, two conditions must be met:
1. You must have allowed T
E
X to call external programs. This is often switched off by default since this
is a security risk (you might, without knowing, run a T
E
X file that calls all sorts of “bad” commands).
To enable this “calling external programs” a command line option must be given to the T
E
X program.
Usually, it is called something like shell-escape or enable-write18. For example, for my pdflatex
the option --shell-escape can be given.
2. You must have installed the gnuplot program and T
E
X must find it when compiling your file.
Unfortunately, these conditions will not always be met. Especially if you pass some source to a coauthor
and the coauthor does not have gnuplot installed, he or she will have trouble compiling your files.
For this reason, TikZ behaves differently when you compile your graphic for the second time:
If upon reaching plot[id=¸id)] function{...} the file ¸prefix)¸id).table already exists and if the
¸prefix)¸id).gnuplot file contains what TikZ thinks that it “should” contain, the .table file is immedi-
ately read without trying to call a gnuplot program. This approach has the following advantages:
1. If you pass a bundle of your .tex file and all .gnuplot and .table files to someone else, that person
can T
E
X the .tex file without having to have gnuplot installed.
2. If the \write18 feature is switched off for security reasons (a good idea), then, upon the first com-
pilation of the .tex file, the .gnuplot will still be generated, but not the .table file. You can then
simply call gnuplot “by hand” for each .gnuplot file, which will produce all necessary .table files.
3. If you change the function that you wish to plot or its domain, TikZ will automatically try to regenerate
the .table file.
4. If, out of laziness, you do not provide an id, the same .gnuplot will be used for different plots, but
this is not a problem since the .table will automatically be regenerated for each plot on-the-fly. Note:
If you intend to share your files with someone else, always use an id, so that the file can by typeset
without having gnuplot installed. Also, having unique ids for each plot will improve compilation
speed since no external programs need to be called, unless it is really necessary.
When you use plot function{¸gnuplot formula)}, the ¸gnuplot formula) must be given in the gnuplot
syntax, whose details are beyond the scope of this manual. Here is the ultra-condensed essence: Use x as
the variable and use the C-syntax for normal plots, use t as the variable for parametric plots. Here are some
examples:
161
x
f(x)
f(x) = x
f(x) = sin x
f(x) =
1
20
e
x
\begin{tikzpicture}[domain=0:4]
\draw[very thin,color=gray] (-0.1,-1.1) grid (3.9,3.9);
\draw[->] (-0.2,0) -- (4.2,0) node[right] {$x$};
\draw[->] (0,-1.2) -- (0,4.2) node[above] {$f(x)$};
\draw[color=red] plot[id=x] function{x} node[right] {$f(x) =x$};
\draw[color=blue] plot[id=sin] function{sin(x)} node[right] {$f(x) = \sin x$};
\draw[color=orange] plot[id=exp] function{0.05*exp(x)} node[right] {$f(x) = \frac{1}{20} \mathrm e^x$};
\end{tikzpicture}
The following options influence the plot:
• samples=¸number) sets the number of samples used in the plot. The default is 25.
• domain=¸start):¸end) sets the domain between which the samples are taken. The default is -5:5.
• parametric=¸true or false) sets whether the plot is a parametric plot. If true, then t must be used
instead of x as the parameter and two comma-separated functions must be given in the ¸gnuplot
formula). An example is the following:
\tikz \draw[scale=0.5,domain=-3.141:3.141,smooth]
plot[parametric,id=parametric-example] function{t*sin(t),t*cos(t)};
• id=¸id) sets the identifier of the current plot. This should be a unique identifier for each plot (though
things will also work if it is not, but not as well, see the explanations above). The ¸id) will be part of
a filename, so it should not contain anything fancy like * or $.
• prefix=¸prefix) is put before each plot file name. The default is \jobname., but if you have many
plots, it might be better to use, say plots/ and have all plots placed in a directory. You have to create
the directory yourself.
• raw gnuplot causes the ¸gnuplot formula) to be passed on to gnuplot without setting up the samples
or the plot operation. Thus, you could write
plot[raw gnuplot,id=raw-example] function{set samples 25; plot sin(x)}
This can be useful for complicated things that need to be passed to gnuplot. However, for really
complicated situations you should create a special external generating gnuplot file and use the file-
syntax to include the table “by hand.”
The following styles influence the plot:
162
• style=every plot This style is installed in each plot, that is, as if you always said
plot[style=every plot,...]
This is most useful for globally setting a prefix for all plots by saying:
\tikzstyle{every plot}=[prefix=plots/]
16.7 Placing Marks on the Plot
As we saw already, it is possible to add marks to a plot using the mark option. When this option is used, a
copy of the plot mark is placed on each point of the plot. Note that the marks are placed after the whole
path has been drawn/filled/shaded. In this respect, they are handled like text nodes.
In detail, the following options govern how marks are drawn:
• mark=¸mark mnemonic) Sets the mark to a mnemonic that has previously been defined using the
\pgfdeclareplotmark. By default, *, +, and x are available, which draw a filled circle, a plus, and a
cross as marks. Many more marks become available when the library pgflibraryplotmarks is loaded.
Section 28.3 lists the available plot marks.
One plot mark is special: the ball plot mark is available only it TikZ. The ball color determines
the balls’s color. Do not use this option with a large number of marks since it will take very long to
render in PostScript.
Option Effect
mark=ball
• mark repeat=¸r) This option tells TikZ that only every rth mark should be drawn.
\tikz \draw plot[mark=x,mark repeat=3,smooth] file {plots/pgfmanual-sine.table};
• mark phase=¸p) This option tells TikZ that the first mark to be draw should be the pth, followed by
the (p +r)th, then the (p + 2r)th, and so on.
\tikz \draw plot[mark=x,mark repeat=3,mark phase=6,smooth] file {plots/pgfmanual-sine.table};
• mark indices=¸list) This option allows you to specify explicitly the indices at which a mark should
be placed. Counting starts with 1. You can use the \foreach syntax, that is, ... can be used.
\tikz \draw plot[mark=x,mark indices={1,4,...,10,11,12,...,16,20},smooth]
file {plots/pgfmanual-sine.table};
163
• mark size=¸dimension) Sets the size of the plot marks. For circular plot marks, ¸dimension) is the
radius, for other plot marks ¸dimension) should be about half the width and height.
This option is not really necessary, since you achieve the same effect by specifying scale=¸factor) as
a local option, where ¸factor) is the quotient of the desired size and the default size. However, using
mark size is a bit faster and more natural.
• mark options=¸options) These options are applied to marks when they are drawn. For example, you
can scale (or otherwise transform) the plot mark or set its color.
\tikz \fill[fill=blue!20]
plot[mark=triangle*,mark options={color=blue,rotate=180}]
file{plots/pgfmanual-sine.table} |- (0,0);
16.8 Smooth Plots, Sharp Plots, and Comb Plots
There are different things the plot operation can do with the points it reads from a file or from the inlined
list of points. By default, it will connect these points by straight lines. However, you can also use options
to change the behavior of plot.
• sharp plot This is the default and causes the points to be connected by straight lines. This option is
included only so that you can “switch back” if you “globally” install, say, smooth.
• smooth This option causes the points on the path to be connected using a smooth curve:
\tikz\draw plot[smooth] file{plots/pgfmanual-sine.table};
Note that the smoothing algorithm is not very intelligent. You will get the best results if the bending
angles are small, that is, less than about 30

and, even more importantly, if the distances between
points are about the same all over the plotting path.
• tension=¸value) This option influences how “tight” the smoothing is. A lower value will result in
sharper corners, a higher value in more “round” curves. A value of 1 results in a circle if four points at
quarter-positions on a circle are given. The default is 0.55. The “correct” value depends on the details
of plot.
164
\begin{tikzpicture}[smooth cycle]
\draw plot[tension=0.2]
coordinates{(0,0) (1,1) (2,0) (1,-1)};
\draw[yshift=-2.25cm] plot[tension=0.5]
coordinates{(0,0) (1,1) (2,0) (1,-1)};
\draw[yshift=-4.5cm] plot[tension=1]
coordinates{(0,0) (1,1) (2,0) (1,-1)};
\end{tikzpicture}
• smooth cycle This option causes the points on the path to be connected using a closed smooth curve.
\tikz[scale=0.5]
\draw plot[smooth cycle] coordinates{(0,0) (1,0) (2,1) (1,2)}
plot coordinates{(0,0) (1,0) (2,1) (1,2)} -- cycle;
• ycomb This option causes the plot operation to interpret the plotting points differently. Instead of
connecting them, for each point of the plot a straight line is added to the path from the x-axis to the
point, resulting in a sort of “comb” or “bar diagram.”
\tikz\draw[ultra thick] plot[ycomb,thin,mark=*] file{plots/pgfmanual-sine.table};
\begin{tikzpicture}[ycomb]
\draw[color=red,line width=6pt]
plot coordinates{(0,1) (.5,1.2) (1,.6) (1.5,.7) (2,.9)};
\draw[color=red!50,line width=4pt,xshift=3pt]
plot coordinates{(0,1.2) (.5,1.3) (1,.5) (1.5,.2) (2,.5)};
\end{tikzpicture}
• xcomb This option works like ycomb except that the bars are horizontal.
\tikz \draw plot[xcomb,mark=x] coordinates{(1,0) (0.8,0.2) (0.6,0.4) (0.2,1)};
• polar comb This option causes a line from the origin to the point to be added to the path for each
plot point.
\tikz \draw plot[polar comb,
mark=pentagon*,mark options={fill=white,draw=red},mark size=4pt]
coordinates {(0:1cm) (30:1.5cm) (160:.5cm) (250:2cm) (-60:.8cm)};
165
• only marks This option causes only marks to be shown; no path segments are added to the actual
path. This can be useful for quickly adding some marks to a path.
\tikz \draw (0,0) sin (1,1) cos (2,0)
plot[only marks,mark=x] coordinates{(0,0) (1,1) (2,0) (3,-1)};
166
17 Transformations
pgf has a powerful transformation mechanism that is similar to the transformation capabilities of metafont.
The present section explains how you can access it in TikZ.
17.1 The Different Coordinate Systems
It is a long process from a coordinate like, say, (1, 2) or (1cm, 5 mathrmpt), to the position a point is finally
placed on the display or paper. In order to find out where the point should go, it is constantly “transformed,”
which means that it is mostly shifted around and possibly rotated, slanted, scaled, and otherwise mutilated.
In detail, (at least) the following transformations are applied to a coordinate like (1, 2) before a point on
the screen is chosen:
1. pgf interprets a coordinate like (1, 2) in its xy-coordinate system as “add the current x-vector once
and the current y-vector twice to obtain the new point.”
2. pgf applies its coordinate transformation matrix to the resulting coordinate. This yields the final
position of the point inside the picture.
3. The backend driver (like dvips or pdftex) adds transformation commands such the coordinate is
shifted to the correct position in T
E
X’s page coordinate system.
4. pdf (or PostScript) apply the canvas transformation matrix to the point, which can once more change
the position on the page.
5. The viewer application or the printer applies the device transformation matrix to transform the coor-
dinate to its final pixel coordinate on the screen or paper.
In reality, the process is even more involved, but the above should give the idea: A point is constantly
transformed by changes of the coordinate system.
In TikZ, you only have access to the first two coordinate systems: The xy-coordinate system and the
coordinate transformation matrix (these will be explained later). pgf also allows you to change the canvas
transformation matrix, but you have to use commands of the core layer directly to do so and you “better
know what you are doing” when you do this. The moment you start modifying the canvas matrix, pgf
immediately looses track of all coordinates and shapes, anchors, and bounding box computations will no
longer work.
17.2 The XY- and XYZ-Coordinate Systems
The first and easiest coordinate systems are pgf’s xy- and xyz-coordinate systems. The idea is very simple:
Whenever you specify a coordinate like (2,3) this means 2v
x
+ 3v
y
, where v
x
is the current x-vector and
v
y
is the current y-vector. Similarly, the coordinate (1,2,3) means v
x
+ 2v
y
+ 3v
z
.
Unlike other packages, pgf does not insist that v
x
actually has a y-component of 0, that is, that it is a
horizontal vector. Instead, the x-vector can point anywhere you want. Naturally, normally you will want
the x-vector to point horizontally.
One undesirable effect of this flexibility is that it is not possible to provide mixed coordinates as in
(1, 2pt). Life is hard.
To change the x-, y-, and z-vectors, you can use the following options:
• x=¸dimension) Sets the x-vector of pgf’s xyz-coordinate system to point ¸dimension) to the right,
that is, to (¸dimension), 0pt). The default is 1cm.
\begin{tikzpicture}
\draw (0,0) -- +(1,0);
\draw[x=2cm,color=red] (0,0.1) -- +(1,0);
\end{tikzpicture}
\tikz \draw[x=1.5cm] (0,0) grid (2,2);
167
The last example shows that the size of steppings in grids, just like all other dimensions, are not
affected by the x-vector. After all, the x-vector is only used to determine the coordinate of the upper
right corner of the grid.
• x=¸coordinate) Sets the x-vector of pgf’s xyz-coordinate system to the specified ¸coordinate). If
¸coordinate) contains a comma, it must be put in braces.
\begin{tikzpicture}
\draw (0,0) -- (1,0);
\draw[x={(2cm,0.5cm)},color=red] (0,0) -- (1,0);
\end{tikzpicture}
You can use this, for example, to exchange the meaning of the x- and y-coordinate.
\begin{tikzpicture}[smooth]
\draw plot coordinates{(1,0) (2,0.5) (3,0) (3,1)};
\draw[x={(0cm,1cm)},y={(1cm,0cm)},color=red]
plot coordinates{(1,0) (2,0.5) (3,0) (3,1)};
\end{tikzpicture}
• y=¸value) Works like the x= option, only if ¸value) is a dimension, the resulting vector points to
(0, ¸value)).
• z=¸value) Works like the z= option, but now a dimension is means the point (¸value), ¸value)).
\begin{tikzpicture}[z=-1cm,->,thick]
\draw[color=red] (0,0,0) -- (1,0,0);
\draw[color=blue] (0,0,0) -- (0,1,0);
\draw[color=orange] (0,0,0) -- (0,0,1);
\end{tikzpicture}
17.3 Coordinate Transformations
pgf and TikZ allow you to specify coordinate transformations. Whenever you specify a coordinate as in
(1,0) or (1cm,1pt) or (30:2cm), this coordinate is first “reduced” to a position of the form “x points to
the right and y points upwards.” For example, (1in,5pt) is reduced to “72
72
100
points to the right and 5
points upwards” and (90:100pt) means “0pt to the right and 100 points upwards.”
The next step is to apply the current coordinate transformation matrix to the coordinate. For example,
the coordinate transformation matrix might currently be set such that it adds a certain constant to the x
value. Also, it might be setup such that it, say, exchanges the x and y value. In general, any “standard”
transformation like translation, rotation, slanting, or scaling or any combination thereof is possible. (Inter-
nally, pgf keeps track of a coordinate transformation matrix very much like the concatenation matrix used
by pdf or PostScript.)
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0,0) rectangle (1,0.5);
\begin{scope}[xshift=1cm]
\draw [red] (0,0) rectangle (1,0.5);
\draw[yshift=1cm] [blue] (0,0) rectangle (1,0.5);
\draw[rotate=30] [orange] (0,0) rectangle (1,0.5);
\end{scope}
\end{tikzpicture}
The most important aspect of the coordinate transformation matrix is that it applies to coordinates only!
In particular, the coordinate transformation has no effect on things like the line width or the dash pattern or
the shading angle. In certain cases, it is not immediately clear whether the coordinate transformation matrix
should apply to a certain dimension. For example, should the coordinate transformation matrix apply to
grids? (It does.) And what about the size of arced corners? (It does not.) The general rule is “If there is
168
no ‘coordinate’ involved, even ‘indirectly,’ the matrix is not applied.” However, sometimes, you simply have
to try or look it up in the documentation whether the matrix will be applied.
Setting the matrix cannot be done directly. Rather, all you can do is to “add” another transformation
to the current matrix. However, all transformations are local to the current T
E
X-group. All transformations
are added using graphic options, which are described below.
Transformations apply immediately when they are encountered “in the middle of a path” and they apply
only to the coordinates on the path following the transformation option.
\tikz \draw (0,0) rectangle (1,0.5) [xshift=2cm] (0,0) rectangle (1,0.5);
A final word of warning: You should refrain from using “aggressive” transformations like a scaling of a
factor of 10000. The reason is that all transformations are done using T
E
X, which has a fairly low accuracy.
Furthermore, in certain situations it is necessary that TikZ inverts the current transformation matrix and
this will fail if the transformation matrix is badly conditioned or even singular (if you do not know what
singular matrices are, you are blessed).
• shift={¸coordinate)} adds the ¸coordinate) to all coordinates.
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0,0) -- (1,1) -- (1,0);
\draw[shift={(1,1)},blue] (0,0) -- (1,1) -- (1,0);
\draw[shift={(30:1cm)},red] (0,0) -- (1,1) -- (1,0);
\end{tikzpicture}
• shift only This option does not take any parameter. Its effect is to cancel all current transformations
except for the shifting. This means that the origin will remain where it is, but any rotation around
the origin or scaling relative to the origin or skewing will no longer have an effect.
This option is useful in situtations where a complicated transformation is used to “get to a position,”
but you then wish to draw something “normal” at this position.
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0,0) -- (1,1) -- (1,0);
\draw[rotate=30,xshift=2cm,blue] (0,0) -- (1,1) -- (1,0);
\draw[rotate=30,xshift=2cm,shift only,red] (0,0) -- (1,1) -- (1,0);
\end{tikzpicture}
• xshift=¸dimension) adds ¸dimension) to the x value of all coordinates.
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0,0) -- (1,1) -- (1,0);
\draw[xshift=2cm,blue] (0,0) -- (1,1) -- (1,0);
\draw[xshift=-10pt,red] (0,0) -- (1,1) -- (1,0);
\end{tikzpicture}
• yshift=¸dimension) adds ¸dimension) to the y value of all coordinates.
• scale=¸factor) multiplies all coordinates by the given ¸factor). The ¸factor) should not be excessively
large in absolute terms or very near to zero.
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0,0) -- (1,1) -- (1,0);
\draw[scale=2,blue] (0,0) -- (1,1) -- (1,0);
\draw[scale=-1,red] (0,0) -- (1,1) -- (1,0);
\end{tikzpicture}
169
• xscale=¸factor) multiplies only the x-value of all coordinates by the given ¸factor).
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0,0) -- (1,1) -- (1,0);
\draw[xscale=2,blue] (0,0) -- (1,1) -- (1,0);
\draw[xscale=-1,red] (0,0) -- (1,1) -- (1,0);
\end{tikzpicture}
• yscale=¸factor) multiplies only the y-value of all coordinates by ¸factor).
• xslant=¸factor) slants the coordinate horizontally by the given ¸factor):
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0,0) -- (1,1) -- (1,0);
\draw[xslant=2,blue] (0,0) -- (1,1) -- (1,0);
\draw[xslant=-1,red] (0,0) -- (1,1) -- (1,0);
\end{tikzpicture}
• yslant=¸factor) slants the coordinate vertically by the given ¸factor):
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0,0) -- (1,1) -- (1,0);
\draw[yslant=2,blue] (0,0) -- (1,1) -- (1,0);
\draw[yslant=-1,red] (0,0) -- (1,1) -- (1,0);
\end{tikzpicture}
• rotate=¸degree) rotates the coordinate system by ¸degree):
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0,0) -- (1,1) -- (1,0);
\draw[rotate=40,blue] (0,0) -- (1,1) -- (1,0);
\draw[rotate=-20,red] (0,0) -- (1,1) -- (1,0);
\end{tikzpicture}
• rotate around={¸degree):¸coordinate)} rotates the coordinate system by ¸degree) around the point
¸coordinate).
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0,0) -- (1,1) -- (1,0);
\draw[rotate around={40:(1,1)},blue] (0,0) -- (1,1) -- (1,0);
\draw[rotate around={-20:(1,1)},red] (0,0) -- (1,1) -- (1,0);
\end{tikzpicture}
• cm={¸a),¸b),¸c),¸d),¸coordinate)} applies the following transformation to all coordinates: Let (x, y)
be the coordinate to be transformed and let ¸coordinate) specify the point (t
x
, t
y
). Then the new
coordinate is given by

a b
c d

(
x
y
) +

t
x
t
y

. Usually, you do not use this option directly.
\begin{tikzpicture}
\draw[style=help lines] (0,0) grid (3,2);
\draw (0,0) -- (1,1) -- (1,0);
\draw[cm={1,1,0,1,(0,0)},blue] (0,0) -- (1,1) -- (1,0);
\draw[cm={0,1,1,0,(1cm,1cm)},red] (0,0) -- (1,1) -- (1,0);
\end{tikzpicture}
170
• reset cm completely resets the coordinate transformation matrix to the identity matrix. This will
destroy not only the transformations applied in the current scope, but also all transformations inherited
from surrounding scopes. Do not use this option, unless you really, really know what you are doing.
171
Part IV
Libraries
by Till Tantau
In this part the library packages are documented. They provide additional predefined graphic objects like
new arrow heads or new plot marks. These are not loaded by default since many users will not need them.
Theoretical
Computer
Science
Theoretical
Computer
Science
Theoretical
Computer
Science
Theoretical
Computer
Science
\tikzstyle{level 1}=
[level distance=4.5mm, trunk, line width=1ex ,sibling angle=60]
\tikzstyle{level 2}=
[level distance=3.5mm, trunk!80!leaf a,line width=.8ex,sibling angle=56]
\tikzstyle{level 3}=
[level distance=2.75mm,trunk!60!leaf a,line width=.6ex,sibling angle=52]
\tikzstyle{level 4}=
[level distance=2mm, trunk!40!leaf a,line width=.4ex,sibling angle=48]
\tikzstyle{level 5}=
[level distance=1mm, trunk!20!leaf a,line width=.3ex,sibling angle=44]
\tikzstyle{level 6}=
[level distance=1.75mm,leaf a, line width=.2ex,sibling angle=40]
\pgfarrowsdeclare{leaf}{leaf}
{\pgfarrowsleftextend{-2pt} \pgfarrowsrightextend{1pt}}
{
\pgfpathmoveto{\pgfpoint{-2pt}{0pt}}
\pgfpatharc{150}{30}{1.8pt}
\pgfpatharc{-30}{-150}{1.8pt}
\pgfusepathqfill
}
\newcommand{\tcslogo}[5]
{
\colorlet{border}{#1}
\colorlet{trunk}{#2}
\colorlet{leaf a}{#3}
\colorlet{leaf b}{#4}
\begin{tikzpicture}
\scriptsize\scshape
\draw[border,line width=1ex,yshift=.3cm,
yscale=1.45,xscale=1.05,looseness=1.42]
(1,0) to [out=90, in=0] (0,1) to [out=180,in=90] (-1,0)
to [out=-90,in=-180] (0,-1) to [out=0, in=-90] (1,0) -- cycle;
\coordinate (root) [grow cyclic,rotate=90]
child {
child [cap=round] foreach \a in {0,1} {
child foreach \b in {0,1} {
child foreach \c in {0,1} {
child foreach \d in {0,1} {
child foreach \leafcolor in {leaf a,leaf b}
{ edge from parent [color=\leafcolor,-#5] }
} } }
} edge from parent [shorten >=-1pt,serif cm-,cap=butt]
};
\node [text centered,text width=2cm,below] at (0pt,-.5ex)
{ \textcolor{border}{T}heoretical \\ \textcolor{border}{C}omputer \\
\textcolor{border}{S}cience };
\end{tikzpicture}
}
\begin{minipage}{3cm}
\tcslogo{green!80!black}{green!25!black}{green}{green!80}{leaf}\\
\tcslogo{green!50!black}{black}{green!80!black}{red!80!green}{leaf}\\
\tcslogo{red!75!black}{red!25!black}{red!75!black}{orange}{leaf}\\
\tcslogo{black!50}{black}{black!50}{black!25}{}
\end{minipage}
172
18 Arrow Tip Library
\usepgflibrary{arrows} % L
A
T
E
X and plain T
E
X and pure pgf
\usepgflibrary[arrows] % ConT
E
Xt and pure pgf
\usetikzlibrary{arrows} % L
A
T
E
X and plain T
E
X when using Tik Z
\usetikzlibrary[arrows] % ConT
E
Xt when using Tik Z
The package defines additional arrow tips, which are described below. See page 322 for the arrows tips
that are defined by default. Note that neither the standard packages nor this package defines an arrow
name containing > or <. These are left for the user to defined as he or she sees fit.
18.1 Triangular Arrow Tips
latex’ yields thick and thin
latex’ reversed yields thick and thin
stealth’ yields thick and thin
stealth’ reversed yields thick and thin
triangle 90 yields thick and thin
triangle 90 reversed yields thick and thin
triangle 60 yields thick and thin
triangle 60 reversed yields thick and thin
triangle 45 yields thick and thin
triangle 45 reversed yields thick and thin
open triangle 90 yields thick and thin
open triangle 90 reversed yields thick and thin
open triangle 60 yields thick and thin
open triangle 60 reversed yields thick and thin
open triangle 45 yields thick and thin
open triangle 45 reversed yields thick and thin
18.2 Barbed Arrow Tips
angle 90 yields thick and thin
angle 90 reversed yields thick and thin
angle 60 yields thick and thin
angle 60 reversed yields thick and thin
angle 45 yields thick and thin
angle 45 reversed yields thick and thin
hooks yields thick and thin
hooks reversed yields thick and thin
18.3 Bracket-Like Arrow Tips
[-] yields thick and thin
]-[ yields thick and thin
(-) yields thick and thin
)-( yields thick and thin
18.4 Circle and Diamond Arrow Tips
o yields thick and thin
* yields thick and thin
diamond yields thick and thin
open diamond yields thick and thin
18.5 Serif-Like Arrow Tips
serif cm yields thick and thin
173
18.6 Partial Arrow Tips
left to yields thick and thin
left to reversed yields thick and thin
right to yields thick and thin
right to reversed yields thick and thin
left hook yields thick and thin
left hook reversed yields thick and thin
right hook yields thick and thin
right hook reversed yields thick and thin
18.7 Line Caps
round cap yields for line width 1ex
butt cap yields for line width 1ex
triangle 90 cap yields for line width 1ex
triangle 90 cap reversed yields for line width 1ex
fast cap yields for line width 1ex
fast cap reversed yields for line width 1ex
174
19 Automata Drawing Library
\usetikzlibrary{automata} % L
A
T
E
X and plain T
E
X
\usetikzlibrary[automata] % ConT
E
Xt
This packages provides shapes and styles for drawing finite state automata and Turing machines.
19.1 Drawing Automata
The automata drawing library is intended to make it easy to draw finite automata and Turing machines.
It does not cover every situation imaginable, but most finite automata and Turing machines found in text
books can be drawn in a nice and convenient fashion using this library.
To draw an automaton, proceed as follows:
1. For each state of the automaton, there should be one node with the option state.
2. To place the states, you can either use absolute positions or relative positions, using options like
above of or right of.
3. Give a unique name to each state node.
4. Accepting and initial states are indicated by adding the options accepting and initial, respectively,
to the state nodes.
5. Once the states are fixed, the edges can be added. For this, the edge operation is most useful. It is,
however, also possible to add edges after each node has been placed.
6. For loops, use the edge [loop] operation.
Let us now see how this works for a real example. Let us consider a nondeterminsitic four state automaton
that checks whether an contains the sequence 0

1 or the sequence 1

0.
q
0 start
q
1
q
2
q
3
0
1
1
0
0
1
\begin{tikzpicture}[shorten >=1pt,node distance=2cm,auto]
\draw[help lines] (0,0) grid (3,2);
\node[state,initial] (q_0) {$q_0$};
\node[state] (q_1) [above right of=q_0] {$q_1$};
\node[state] (q_2) [below right of=q_0] {$q_2$};
\node[state,accepting](q_3) [below right of=q_1] {$q_3$};
\path[->] (q_0) edge node {0} (q_1)
edge node [swap] {1} (q_2)
(q_1) edge node {1} (q_3)
edge [loop above] node {0} ()
(q_2) edge node [swap] {0} (q_3)
edge [loop below] node {1} ();
\end{tikzpicture}
175
19.2 States With and Without Output
The state style actually just “selects” a default underlying style. Thus, you can define multiple new
complicated state style and then simply set the state style to your given style to get the desired kind of
styles.
By default, the following state styles are defined:
• style=state without output This node style causes nodes to be drawn circles. Also, this style calls
every state.
• style=state with output This node style causes nodes to be drawn as split circles, that is, using the
circle split shape. In the upper part of the shape you have the name of the style, in the lower part
the output is placed. To specify the output, use the command \nodepart{lower} inside the node.
This style also calls every state.
q
0
q
1
00
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\node[state without output] {$q_0$};
\node[state with output] at (2,0) {$q_1$ \nodepart{lower} $00$};
\end{tikzpicture}
• style=state This style is set to state without output by default. You should redefine it to some-
thing else, if you wish to use states of a different nature.
q
0
11
q
1
00
\begin{tikzpicture}
\tikzstyle{state}=[state with output]
\node[state] {$q_0$ \nodepart{lower} $11$};
\node[state] at (2,0) {$q_1$ \nodepart{lower} $00$};
\end{tikzpicture}
• style=every state This style is used by state with output and also by state without output.
By default, it does nothing, but you can use it to make your state look more fancy:
q
0 start
q
1
q
2
0
1
0
1
\begin{tikzpicture}[shorten >=1pt,node distance=2cm,>=stealth’]
\tikzstyle{every state}=[draw=blue!50,very thick,fill=blue!20]
\node[state,initial] (q_0) {$q_0$};
\node[state] (q_1) [above right of=q_0] {$q_1$};
\node[state] (q_2) [below right of=q_0] {$q_2$};
\path[->] (q_0) edge node [above left] {0} (q_1)
edge node [below left] {1} (q_2)
(q_1) edge [loop above] node {0} ()
(q_2) edge [loop below] node {1} ();
\end{tikzpicture}
19.3 Initial and Accepting States
The styles initial and accepting are similar to the state style as they also just select an “underlying”
style, which installs the actual settings for initial and accepting states.
For initial states, there is only one predefined way of drawing them, so the two-stage mechanism is not
really necessary, but, perhaps, I will find another way of drawing them in the literature some time.
Let us start with the initial states.
• style=initial This style simply selects initial by arrow.
176
• style=initial by arrow This style causes an arrow and, possibly, some text to be added to the node.
The arrow points from the text to the node. The node text and the direction and the distance can be
set using the following options:
– initial text=¸text) sets the text to be used. Use an empty text to suppress all text. The default
is start.
– initial where=¸direction) set the place where the text should be shown. Allowed values are
above, below, left, and right.
– intial distance=¸distance) is the length of the arrow leading from the text to the state node.
• style=every initial by arrow This style is executed at the beginning of every path that contains
the arrow and the text. You can use it to, say, make the text red or whatever.
q
0 start
\begin{tikzpicture}
\tikzstyle{every initial by arrow}=[text=red,->>]
\node[state,initial,initial distance=2cm] {$q_0$};
\end{tikzpicture}
• style=initial above is a shorthand for initial by arrow,initial where=above
• style=initial below works similarly to the previous option.
• style=initial left works similarly to the previous option.
• style=initial right works similarly to the previous option.
For the accepting states, the sitation is similar: There is also an accepting style that selects the way
accepting states are rendered. However, there are now two options: First, accepting by arrow, which works
the same way as initial by arrow, only with the direction of arrow reversed, and accepting by double,
where accepting states get a double line around them.
• style=accepting This style selects accepting by double by default. You can replace this by the
style accepting by arrow to get accepting states with an arrow leaving them.
• style=accepting by double Specifies that the node should get a double line around it.
• style=accepting by arrow This style causes an arrow and, possibly, some text to be added to the
node. The arrow points to the text from the node.
– accepting text=¸text) sets the text to be used. Use an empty text to suppress all text. This is
the default.
– accepting where=¸direction) set the place where the text should be shown. Allowed values are
above, below, left, and right.
– intial distance=¸distance) is the length of the arrow.
q
0
q
1
q
2
q
3
0
1
1
0
0
1
177
\begin{tikzpicture}
[shorten >=1pt,node distance=2cm,>=stealth’,initial text=]
\tikzstyle{every state}=[draw=blue!50,very thick,fill=blue!20]
\tikzstyle{accepting}=[accepting by arrow]
\node[state,initial] (q_0) {$q_0$};
\node[state] (q_1) [above right of=q_0] {$q_1$};
\node[state] (q_2) [below right of=q_0] {$q_2$};
\node[state,accepting](q_3) [below right of=q_1] {$q_3$};
\path[->] (q_0) edge node [above left] {0} (q_1)
edge node [below left] {1} (q_2)
(q_1) edge node [above right] {1} (q_3)
edge [loop above] node {0} ()
(q_2) edge node [below right] {0} (q_3)
edge [loop below] node {1} ();
\end{tikzpicture}
• style=every accepting by arrow This style is executed at the beginning of every path that contains
the arrow and the text.
• style=accepting above is a shorthand for accepting by arrow,accepting where=above
• style=accepting below works similarly to the previous option.
• style=accepting left works similarly to the previous option.
• style=accepting right works similarly to the previous option.
19.4 Examples
In the following example, we once more typeset the automaton presented in the previous sections. This
time, we use the following rule for accepting/initial state: Initial states are red, accepting states are green,
and normal states are orange. Then, we must find a path from a red state to a green state.
q
0
q
1
q
2
q
3
0
1
1
0
0
1
\begin{tikzpicture}[shorten >=1pt,node distance=2cm,>=stealth’,thick]
\tikzstyle{every state}=[fill,draw=none,orange,text=white]
\tikzstyle{accepting}=[green!50!black,text=white]
\tikzstyle{initial}= [red,text=white]
\node[state,initial] (q_0) {$q_0$};
\node[state] (q_1) [above right of=q_0] {$q_1$};
\node[state] (q_2) [below right of=q_0] {$q_2$};
\node[state,accepting](q_3) [below right of=q_1] {$q_3$};
\path[->] (q_0) edge node [above left] {0} (q_1)
edge node [below left] {1} (q_2)
(q_1) edge node [above right] {1} (q_3)
edge [loop above] node {0} ()
(q_2) edge node [below right] {0} (q_3)
edge [loop below] node {1} ();
\end{tikzpicture}
The next example is the current candidate for the five-state busiest beaver:
178
q
a start
q
b
q
d
q
c
q
e
0,1,L
1,1,R
1,1,L
0,1,L
0,1,L
1,0,R
1,1,R
0,1,R
1,0,R
\begin{tikzpicture}[->,>=stealth’,shorten >=1pt,%
auto,node distance=2cm,semithick,
inner sep=2pt,bend angle=45]
\node[initial,state] (A) {$q_a$};
\node[state] (B) [above right of=A] {$q_b$};
\node[state] (D) [below right of=A] {$q_d$};
\node[state] (C) [below right of=B] {$q_c$};
\node[state] (E) [below of=D] {$q_e$};
\tikzstyle{every node}=[font=\footnotesize]
\path (A) edge node {0,1,L} (B)
edge node {1,1,R} (C)
(B) edge [loop above] node {1,1,L} (B)
edge node {0,1,L} (C)
(C) edge node {0,1,L} (D)
edge [bend left] node {1,0,R} (E)
(D) edge [loop below] node {1,1,R} (D)
edge node {0,1,R} (A)
(E) edge [bend left] node {1,0,R} (A);
\end{tikzpicture}
179
20 Background Library
\usetikzlibrary{backgrounds} % L
A
T
E
X and plain T
E
X
\usetikzlibrary[backgrounds] % ConT
E
Xt
This library defines “backgrounds” for pictures. This does not refer to background pictures, but rather
to frames drawn around and behind pictures. For example, this package allows you to just add the
framed option to a picture to get a rectangular box around your picture or gridded to put a grid
behind your picture.
When this package is loaded, the following styles become available:
• style=show background rectangle This style causes a rectangle to be drawn behind your graphic.
This style option must be given to the {tikzpicture} environment or to the \tikz command.
\begin{tikzpicture}[show background rectangle]
\draw (0,0) ellipse (10mm and 5mm);
\end{tikzpicture}
The size of the background rectangle is determined as follows: We start with the bounding box of the
picture. Then, a certain separator distance is added on the sides. This distance can be different for
the x- and y-directions and can be set using the following options:
– inner frame xsep=¸dimension) Sets the additional horizontal separator distance for the back-
ground rectangle. The default is 1ex.
– inner frame ysep=¸dimension) Same for the vertical separator distance.
– inner frame sep=¸dimension) sets the horizontal and vertical separator distances simultane-
ously.
The following two styles make setting the inner separator a bit easier to remember:
– style=tight background Sets the inner frame separator to 0pt. The background rectangle will
have the size of the bounding box.
– style=loose background Sets the inner frame separator to 2ex.
You can influence how the background rectangle is rendered by setting the following style:
– style=background rectangle This style dictates how the background rectangle is drawn or filled.
By default this style is set to draw, which causes the path of the background rectangle to be drawn
in the usual way. Setting this style to, say, fill=blue!20 causes a light blue background to be
added to the picture. You can also use more fancy settings as shown in the following example:
\tikzstyle{background rectangle}=
[double,ultra thick,draw=red,top color=blue,rounded corners]
\begin{tikzpicture}[show background rectangle]
\draw (0,0) ellipse (10mm and 5mm);
\end{tikzpicture}
Naturally, no one in their right mind would use the above, but here is a nice background:
\tikzstyle{background rectangle}=
[draw=blue!50,fill=blue!20,rounded corners=1ex]
\begin{tikzpicture}[show background rectangle]
\draw (0,0) ellipse (10mm and 5mm);
\end{tikzpicture}
• style=framed This is a shorthand for show background rectangle.
• style=show background grid This style behaves similarly to the show background rectangle style,
but it will not use a rectangle path, but a grid. The lower left and upper right corner of the grid is
computed in the same way as for the background rectangle:
180
\begin{tikzpicture}[show background grid]
\draw (0,0) ellipse (10mm and 5mm);
\end{tikzpicture}
You can influence the background grid by setting the following style:
– style=background grid This style dictates how the background grid path is drawn. The default
is draw,help lines.
\tikzstyle{background grid}=[thick,draw=red,step=.5cm]
\begin{tikzpicture}[show background grid]
\draw (0,0) ellipse (10mm and 5mm);
\end{tikzpicture}
This option can be combined with the framed option (use the framed option first):
\tikzstyle{background grid}=[thick,draw=red,step=.5cm]
\tikzstyle{background rectangle}=[rounded corners,fill=yellow]
\begin{tikzpicture}[framed,gridded]
\draw (0,0) ellipse (10mm and 5mm);
\end{tikzpicture}
– style=gridded This is a shorthand for show background grid.
• style=show background top This style causes a single line to be drawn at the top of the background
rectangle. Normally, the line coincides exactly with the top line of the background rectangle:
\tikzstyle{background rectangle}=[fill=yellow]
\begin{tikzpicture}[framed,show background top]
\draw (0,0) ellipse (10mm and 5mm);
\end{tikzpicture}
The following option allows you to lengthen (or shorten) the line:
– outer frame xsep=¸dimension) The ¸dimension) is added at the left and right side of the line.
\tikzstyle{background rectangle}=[fill=yellow]
\begin{tikzpicture}
[framed,show background top,outer frame xsep=1ex]
\draw (0,0) ellipse (10mm and 5mm);
\end{tikzpicture}
– outer frame ysep=¸dimension) This option does not apply to the top line, but to the left and
right lines, see below.
– outer frame sep=¸dimension) Sets both the x- and y-separation.
\tikzstyle{background rectangle}=[fill=blue!20]
\begin{tikzpicture}
[outer frame sep=1ex,%
show background top,%
show background bottom,%
show background left,%
show background right]
\draw (0,0) ellipse (10mm and 5mm);
\end{tikzpicture}
You can influence how the line is drawn grid by setting the following style:
– style=background top Default is draw.
\tikzstyle{background rectangle}=[fill=blue!20]
\tikzstyle{background top}=[draw=blue!50,line width=1ex]
\begin{tikzpicture}[framed,show background top]
\draw (0,0) ellipse (10mm and 5mm);
\end{tikzpicture}
181
• style=show background bottom works like the style for the top line.
• style=show background left works like the style for the top line.
• style=show background right works like the style for the top line.
182
21 Calendar Library
\usetikzlibrary{calendar} % L
A
T
E
X and plain T
E
X
\usetikzlibrary[calendar] % ConT
E
Xt
The library defines the \calendar command, which can be used to typeset calendars. The command
relies on the \pgfcalendar command from the pgfcalendar package, which is loaded automatically.
The \calendar command is quite configurable, allowing you to produce all kinds of different calendars.
21.1 Calendar Command
The core command for creating calendars in TikZ is the \calendar command. It is available only inside
{tikzpicture} environments (similar to, say, the \draw command).
\calendar¸calendar specification);
The syntax for this command is similar to commands like \node or \matrix. However, it has its
complete own parser and only those commands described in the following will be recognized, nothing
else. Note, furthermore, that a ¸calendar specification) is not a path specification, indeed, no path is
created for the calendar.
The specification syntax. The ¸calendar specification) must be a sequence of elements, each of which
has one of the following structures:
• [¸options)]
You provide ¸options) in square brackets as in [red,draw=none]. These ¸options) can be any
TikZ option and they apply to the whole calendar. You can provide this element multiple times,
the effect accumulates.
• (¸name))
This has the same effect as saying [name=¸name)]. The effect of providing a ¸name) is explained
later. Note alreadys that a calendar is not a node and the ¸name) is not the name of a node.
• at (¸coordinate))
This has the same effect as saying [at=(¸coordinate))].
• if (¸date condition)) ¸options or commands)else¸else options or commands)
The effect of such an if is explained later.
At the beginning of every calendar, the following style is used:
• style=every calendar This style is empty be default.
The date range. The overall effect of the \calendar command is to execute code for each day of a
range of dates. This range of dates is set using the following option:
• dates=¸start date) to ¸end date) This option specifies the date range. Both the start and end
date are specified as described on page 250. In short: You can provide ISO-format type dates like
2006-01-02, you can replace the day of month by last to refer to the last day of a month (so
2006-02-last is the same as 2006-02-28), and you can add a plus sign followed by a number to
specify an offset (so 2006-01-01+-1 is the same as 2005-12-31).
It will be useful to fix two pieces of terminology for the following descriptions: The \calendar command
iterates over the dates in the range. The current date refers to the current date the command is
processing as it iterates over the dates. For each current date code is executed, which will be called the
current date code. The current date code consists of different parts, to be detailed later.
The central part of the current date code is the execution of the code \tikzdaycode. By default, this
code simply produces a node whose text is set to the day of month. This means that unless further
action is taken, all days of a calendar will be put on top of each other! To avoid this, you must modify the
current date code to shift days around appropriately. Predefined arrangements like day list downward
or week list do this for you, but you can define arrangements yourself. Since defining an arrangement
is a bit tricky, it is explained only later on. For the time being, let us use a predefined arrangement to
produce our first calendar:
183
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
\tikz \calendar[dates=2000-01-01 to 2000-01-31,week list];
Changing the spacing. In the above calendar, the spacing between the days is determined by the
numerous options. Most arrangement do not use all of these options, but only those that apply naturally.
• day xshift=¸dimension) specifies the horizontal shift between days. This is not the gap between
days, but the shift between the anchors of their nodes. The default is 3.5ex.
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
\tikz \calendar[dates=2000-01-01 to 2000-01-31,week list,day xshift=3ex];
• day yshift=¸dimension) specifies the vertical shift between days. Again, this is the shift between
the anchors of their nodes. The default is 3ex.
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
\tikz \calendar[dates=2000-01-01 to 2000-01-31,week list,day yshift=2ex];
• month xshift=¸dimension) specifies an additional horizontal shift between different months.
• month yshift=¸dimension) specifies an additional vertical shift between different months.
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31 1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29
\tikz \calendar[dates=2000-01-01 to 2000-02-last,week list,
month yshift=0pt];
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29
\tikz \calendar[dates=2000-01-01 to 2000-02-last,week list,
month yshift=1cm];
184
Changing the position of the calendar. The calendar is placed in such a way that, normally, the
anchor of the first day label is at the origin. This can be changed by using the at option. When you
say at={(1,1)}, this anchor of the first day will lie at coordinate (1, 1).
In general, arrangements will not always place the anchor of the first day at the origin. Sometimes,
additional spacing rules get in the way. There are different ways of addressing this problem: First, you
can just ignore it. Since calendars are often placed in their own {tikzpicture} and since their size if
computed automatically, the exact position of the origin often does not matter at all. Second, you can
put the calendar inside a node as in ...node {\tikz \calendar...}. This allows you to position the
node in the normal ways using the node’s anchors. Third, you can be very clever and use a single-cell
matrix. The advantage is that a matrix allows you to provide any anchor of any node inside the matrix
as an anchor for the whole matrix. For example, the following calendar is placed in such a way the
center of 2000-01-20 lies on the position (2, 2):
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\matrix [anchor=cal-2000-01-20.center] at (2,2)
{ \calendar(cal)[dates=2000-01-01 to 2000-01-31,week list]; \\};
\end{tikzpicture}
Unfortunately, the matrix-base positions, which is the cleanest way, isnot as portable as the other
approaches (it currently does not work with the svg backend for instance).
Changing the appearance of days. As mentioned before, each day in the above calendar is produced
by an execution of the \tikzdaycode. Each time this code is executed, the coordinate system will have
been setup appropriately to place the day of the month correctly. You can change both the code and
its appearance using the following options.
• day code=¸code) This option allows you to change the code that is executed for each day. The
default is to create a node with an appropriate name, but you can change this:
\tikz \calendar[dates=2000-01-01 to 2000-01-31,week list,
day code={\fill[blue] (0,0) circle (2pt);}];
The default code is the following:
\node[name=\pgfcalendarsuggestedname,every day]{\tikzdaytext};
The first part causes the day nodes to be accessible via the following names: If ¸name) is the
name given to the calendar via a name= option or via the specification element (¸name)), then
\pgfcalendarsuggestedname will expand to ¸name)-¸date), where ¸date) is the date of the day
that is currently being processed in ISO format .
For example, if January 1, 2006 is being processed and the calendar has been named mycal, then
the node containg the 1 for this date will be names mycal-2006-01-01. You can later reference
this node.
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
\begin{tikzpicture}
\calendar (mycal) [dates=2000-01-01 to 2000-01-31,week list];
\draw[red] (mycal-2000-01-20) circle (4pt);
\end{tikzpicture}
185
• day text=¸text) This option changes the setting of the \tikzdaytext. By default, this macro
simply yields the current day of month, but you can change it arbitrarily. Here is a silly example:
x x
x x x x x x x
x x x x x x x
x x x x x x x
x x x x x x x
x
\tikz \calendar[dates=2000-01-01 to 2000-01-31,week list,
day text=x];
More useful examples are based on using the \% command. This command is redefined inside a
\pgfcalendar to mean the same as \pgfcalendarshorthand. (The original meaning of \% is lost
inside the calendar, you need to save if before the calendar if you really need it.)
The \% inserts the current day/month/year/day of week in a certain format into the text. The
first letter following the \% selects the type (permisslbe values are d, m, y, w), the second letter
specifies how the value should be displayed (- means numerically, = means numerically with leading
space, 0 means numerically with leading zeros, t means textual, and . means textual, abbrevi-
ated). For example \%d0 gives the day with a leading zero (for more details see the description of
\pgfcalendarshorthand on page 255).
Let us redefine the day text so that it yields the day with a leading zero:
01 02
03 04 05 06 07 08 09
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
\tikz \calendar[dates=2000-01-01 to 2000-01-31,week list,
day text=\%d0];
• style=every day This style is executed by the default node code for each day. The default setting
is
anchor=base east
The every day style is useful for changing the way days look. For example, let us make all days
red:
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
\tikzstyle{every day}+=[red]
\tikz \calendar[dates=2000-01-01 to 2000-01-31,week list];
Changing the appearance of month and year labels. In addition to the days of a calendar, labels
for the months and even years (for really long calendars) can be added. These labels are only added once
per month or year and this is not done by default. Rather, special styles starting with month label
place these labels and make them visible:
186
January
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
February
1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29
\tikz \calendar[dates=2000-01-01 to 2000-02-last,week list,
month label above centered];
The following options change the appearance of the month and year label:
• month code=¸code) This option allows you to specify what the macro \tikzmonthcode should
expand to.
By default, the \tikzmonthcode it is set to
\node[every month]{\tikzmonthtext};
Note that this node is not named by default.
• month text=¸text) This option allows you to change the macro \tikzmonthtext. By default, the
month text is a long textual presentation of the current month being typeset.
January 2000
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
\tikz \calendar[dates=2000-01-01 to 2000-01-31,week list,
month label above centered,
month text=\textcolor{red}{\%mt} \%y-];
• style=every month This style, which is empty by default, can be used to change the appearance
of month labels.
• year code=¸code) Works like month code, only for years.
• year text=¸text) Works like month text, only for years.
• style=every year Works like every month, only for years.
Date ifs. Much of the power of the \calendar command comes from the use of conditionals. There are
two equivalent way of specifying such a conditional. First, you can add the text if (¸conditions)) ¸code
or options) to your ¸calendar specification), possibly followed by else¸else code or options). You can
have multiple such conditionals (but you cannot nest them in this simple manner). The second way is
to use the following option:
• if=(¸coditions))¸code or options)else¸else code or options) This option has the same effect as
giving a corresponding if in the ¸calendar specification). The option is mostly useful for use in the
every calendar style, where you cannot provide if conditionals otherwise.
Now, regardless of how you specify a conditional, it has the following effect (individually and indepen-
dently for each date in the calendar):
1. It is checked whether the current date is one of the possibilities listed in ¸coditions). An example
of such a condition is Sunday. Thus, when you write if (Saturday,Sunday) {foo}, then foo will
be executed for every day in the calendar that is a Saturday or a Sunday.
187
The command \ifdate and, thereby, \pgfcalendarifdate are used to evaluate the ¸conditions),
see page 251 for a complete list of possible tests. The most useful tests are: Tests like Monday
and so on, workday for the days Monday to Friday, weekend for Saturday and Sunday, equals for
testing whether the current date equals a given date, at least and at least for comparing the
current date with a given date.
2. If the date passes the check, the ¸code or options) is evaluated in a manner to be described in a
moment; if the date fails, the ¸else code or options) is evaluated, if present.
The ¸code or options) can either be some code. This is indicated by surrounding the code with
curly braces. It can also be a list of TikZ options. This is indicated by surrounding the options
with square brackets. For example in the date test if (Sunday) {\draw...} else {\fill...}
there are two pieces of code involved. By comparison, if (Sunday) [red] else [green] involves
two options.
If ¸code or options) is code, it is simply executed (for the current day). If it is a list of options,
these options are passed to a scope surrounding the current date.
Let us now have a look at some examples. First, we use a conditional to make all Sundays red.
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
\tikz
\calendar
[dates=2000-01-01 to 2000-01-31,week list]
if (Sunday) [red];
Next, let us do something on a specific date:
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
\tikz
\calendar
[dates=2000-01-01 to 2000-01-31,week list]
if (Sunday) [red]
if (equals=2000-01-20) {\draw (0,0) circle (8pt);};
You might wonder why the circle seems to be “off” the date. Actually, it is centered on the date, it is
just that the date label uses the base east anchor, which shifts the label up and right. To overcome
this problem we can change the anchor:
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
\tikzstyle{every day}=[anchor=mid]
\tikz
\calendar
[dates=2000-01-01 to 2000-01-31,week list]
if (Sunday) [red]
if (equals=2000-01-20) {\draw (0,0) circle (8pt);};
However, the single day dates are now no longer aligned correctly. For this, we can change the day text
to \%d=, which adds a space at the beginning of single day text.
In the following, more technical information is covered. Most readers may wish to skip it.
The current date code. As mentioned earlier, for each date in the calendar the current date code is
executed. It is the job of this code to shift around date nodes, to render the date nodes, to draw the
month labels and to do all other stuff that is necessary to draw a calendar.
The current daet code consists of the following parts, in this order:
1. The before-scope code.
2. A scope is opened.
3. The at-begin-scope code.
188
4. All date-ifs from the ¸calendar specification) are executed.
5. The at-end-scope code.
6. The scope is closed.
7. The after-scope code.
All of the codes mentioned above can be changed using appropriate options, see below. In case you
wonder why so many are needed, the reason is that the current date code as a whole is not surrounded
by a scope or T
E
X group. This means that code executed in the before-scope code and in the after-scope
code has an effect on all following days. For example, if the after-scope code modifies the transformation
matrix by shifting everything downward, all following days will be shifted downward. If each day does
this, you get a list of days, one below the other.
However, you do not always want code to have an effect on everything that follows. For instance, if a
day has the date-if if (Sunday) [red], we only want this Sunday to red, not all following days also.
Similarly, sometimes it is easier to compute the position of a day relative to a fixed origin and we do
not want any modifications of the transformation matrix to have an effect outside the scope.
By cleverly adjusting the different codes, all sorts of different day arrangements are possible.
• execute before day scope=¸code) This code is executed before everything else for each date.
Multiple calls of this option have an accumulative effect. Thus, if you use this option twice, the
code from the first use is used first for each day, followed by the code given the second time.
• execute at begin day scope=¸code) This code is execute before everything else inside the scope
of the current date. Again, the effect is accumulative.
• execute at end day scope=¸code) This code is executed just before the day scope is closed. The
effect is also accumulative, however, in reverse order. This is useful to pair, say, \scope and
\endscope commands in at-begin- and at-end-code.
• execute after day scope=¸code) This is executed at the very end of the current date, outside
the scope. The accumulation is also in reverse.
In the rest of the following subsections we have a look at how the different scope codes can be used to
create different calendar arrangements.
21.1.1 Creating a Simple List of Days
We start with a list the days of the calendar, one day below the other. For this, we simply shift the coordinate
system downward at the end of the code for each day. This shift must be outside the day scope as we want
day shifts to accumulate. Thus, we use the following code:
1
2
3
4
5
6
7
8
\tikz
\calendar [dates=2000-01-01 to 2000-01-08,
execute after day scope=
{\pgftransformyshift{-1em}}];
Clearly, we can use this approach to create day lists going up, down, right, left, or even diagonally.
21.1.2 Adding a Month Label
We now want to add a month label to the left of the beginning of each month. The idea is to do two things:
1. We add code that is executed only on the first of each month.
2. The code is executed before the actual day is rendered. This ensures that options applying to the days
do not affect the month rendering.
We have two options where we should add the month code: Either we add it at the beginning of the day
scope or before. Either will work fine, but it might be safer to put the code inside the scope to ensure that
settings to not inadventerdly “leak outside.”
189
January 1
2
3
4
5
6
7
8
\tikz
\calendar
[dates=2000-01-01 to 2000-01-08,
execute after day scope={\pgftransformyshift{-1em}},
execute at begin day scope=
{\ifdate{day of month=1}{\tikzmonthcode}{}},
set style={{every month}+=[anchor=base east,xshift=-2em]}];
In the above code we used the \ifdate{¸condition)}{¸then code)}{¸else code)} command, which
is described on page 253 in detail and which has much the same effect as if (¸condition)){¸then
code)} else {¸else code)}, but works in normal code.
21.1.3 Creating a Week List Arrangement
Let us now address a more complicated arrangement: A week list. In this arrangement there is line for each
week. The horizontal placement of the days is thus that all Mondays lie below each other, likewise for all
Tuesdays, and so on.
In order to typeset this arrangement, we can use the following approach: The origin of the coordinate
system rests at the anchor for the Monday of each week. That means that at the end of each week the origin
is moved downward one line. On all other days, the origin at the end of the day code is the same as at
the beginning. To position each day correctly, we use code inside and at the beginning of the day scope to
horizontally shift the day according to its day of week.
1 2
3 4 5 6 7 8 9
10111213141516
17181920
\tikz
\calendar
[dates=2000-01-01 to 2000-01-20,
% each day is shifted right according to the day of week
execute at begin day scope=
{\pgftransformxshift{\pgfcalendarcurrentweekday em}},
% after each week, the origin is shifted downward:
execute after day scope=
{\ifdate{Sunday}{\pgftransformyshift{-1em}}{}}];
21.1.4 Creating a Month List Arrangement
For another example, let us create an arrangment that contains one line for each month. This is easy enough
to do as for weeks, unless we add the following requirement: Again, we want all days in a column to have
the same day of week. Since months start on different days of week, this means that each row has to have
an individual offset.
One possible way is to use the following approach: After each month (or at the beginning of each month)
we advance the vertical position of the offset by one line. For horizontal placement, inside the day scope we
locally shift the day by its day of month. Furthermore, we must additionally shift the day to ensure that
the first day of the month lies on the correct day of week column. For this, we remember this day of week
the first time we see it.
1 2 3 4 5 6 7 8 910111213141516171819202122232425262728293031
1 2 3 4 5 6 7 8 91011121314151617181920212223242526272829
190
\newcount\mycount
\tikz
\calendar
[dates=2000-01-01 to 2000-02-last,
execute before day scope=
{
\ifdate{day of month=1} {
% Remember the weekday of first day of month
\mycount=\pgfcalendarcurrentweekday
% Shift downward
\pgftransformyshift{-1em}
}{}
},
execute at begin day scope=
{
% each day is shifted right according to the day of month
\pgftransformxshift{\pgfcalendarcurrentday em}
% and additionally according to the weekday of the first
\pgftransformxshift{\the\mycount em}
}];
21.2 Arrangements
An arrangement specifies how the days of calendar are arranged on the page. The calendar library defines
a number of predefined arrangements.
We start with arrangements in which the days are listed in a long line.
• style=day list downward This style causes the days of a month to be typeset one below the other.
The shift between days is given by day yshift. Between month an additional shift of month yshift
is added.
28
29
30
31
1
2
3
\tikz
\calendar [dates=2000-01-28 to 2000-02-03,
day list downward,month yshift=1em];
• style=day list upward works as above, only the list grows upward instead of downward.
28
29
30
31
1
2
3
\tikz
\calendar [dates=2000-01-28 to 2000-02-03,
day list upward,month yshift=1em];
• style=day list right This style also works as before, but the list of days grows to the right. Instead
of day yshift and month yshift, the values of day xshift and month xshift are used.
28 29 30 31 1 2 3
\tikz
\calendar [dates=2000-01-28 to 2000-02-03,
day list right,month xshift=1em];
• style=day list left as above, but the list grows left.
191
The next arrangement lists days weekwise.
• style=week list This style creates one row for each week in the range. The value of day xshift is
used for the distance between days in each week row, the value of day yshift is used for the distance
between rows. In both cases, “distance” refers to the distance between the anchors of the nodes of the
days (or, more generally, the distance between the origins of the little pictures created for each day).
The days inside each week are shifted such that Monday is always at the first position (to change
this, you need to copy and then modify the code appropriately). If the date range does not start on a
Monday, the first line will not start in the first column, but rather in the column appropriate for the
first date in the range.
At the beginning of each month (except for the first month in the range) an additional vertical space
of month yshift is added. If this is set to 0pt you get a continuous list of days.
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29
\tikz
\calendar [dates=2000-01-01 to 2000-02-last,week list];
1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31 1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29
\tikz
\calendar [dates=2000-01-01 to 2000-02-last,week list,
month yshift=0pt];
The following arrangement gives a very compact view of a whole year.
• style=month list In this arrangement there is a row for each month. As for the week list, the
day xshift is used for the horizontal distance. For the vertical shft, month yshift is used.
In each row, all days of the month are listed alongside each other. However, it is once more ensured
that days in each column lie on the same day of week. Thus, the very first column contains only
Mondays. If a month does not start with a Monday, its days are shifted to the right such that the days
lie on the correct columns.
January 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
February 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
March 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
April 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
May 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
June 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
July 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
August 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
September 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
October 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
November 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
December 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
192
\sffamily\scriptsize
\tikz
\calendar [dates=2000-01-01 to 2000-12-31,
month list,month label left,month yshift=1.25em]
if (Sunday) [black!50];
21.3 Month Labels
For many calendars you may wish to add a labels to each month. We have already covered how month nodes
are created and rendered in the description of the \calendar command: use month text, every month, and
also month code (if necessary) to change the appearance of the month labels.
What we have not yet covered is where these labels are placed. By default, they are not placed at all as
there is no good “default position” for them. Instead, you can use one of the following options to specify a
position for the labels:
• style=month label left Places the month label to the left of the first day of the month. (For
week list and month list where a month does not start on a Monday, the position is chosen “as if”
the month had started on a Monday – which is usually exactly what you want.)
28
29
30
31
February 1
2
3
\tikz
\calendar [dates=2000-01-28 to 2000-02-03,
day list downward,month yshift=1em,
month label left];
• style=month label left vertical This style works like the above style, only the label is rotated
counterclockwise by 90 degrees.
28
29
30
31
F
e
b
r
u
a
r
y 1
2
3
\tikz
\calendar [dates=2000-01-28 to 2000-02-03,
day list downward,month yshift=1em,
month label left vertical];
• style=month label right This style places the month label to the right of the row in which the first
day of the month lies. This means that for a day list the label is to the right of the first day, for a
week list it is to the right of the first week, and for a month list it is to the right of the whole month.
28
29
30
31
February 1
2
3
\tikz
\calendar [dates=2000-01-28 to 2000-02-03,
day list downward,month yshift=1em,
month label right];
• style=month label right vertical as above, only the label is rotated clockwise by 90 degrees.
193
28
29
30
31
F
e
b
r
u
a
r
y
1
2
3
\tikz
\calendar [dates=2000-01-28 to 2000-02-03,
day list downward,month yshift=1em,
month label right vertical];
• style=month label above left This style places the month label above of the row of the first day,
flushed left to the leftmost column. The amount by which the label is raised is fixed to 1.25em; use
the yshift option with the month node to modify this.
28 29 30 31
February
1 2 3
\tikz
\calendar [dates=2000-01-28 to 2000-02-03,
day list right,month xshift=1em,
month label above left];
20 21 22 23
24 25 26 27 28 29 30
31
February
1 2 3 4 5 6
7 8 9 10
\tikz
\calendar [dates=2000-01-20 to 2000-02-10,
week list,month label above left];
• style=month label above centered works as above, only the label is centered above the row con-
taining the first day.
February
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
\tikz
\calendar [dates=2000-02-01 to 2000-02-last,
day list right,month label above centered];
20 21 22 23
24 25 26 27 28 29 30
31
February
1 2 3 4 5 6
7 8 9 10
\tikz
\calendar [dates=2000-01-20 to 2000-02-10,
week list,month label above centered];
• style=month label above right works as above, but flushed right
194
20 21 22 23
24 25 26 27 28 29 30
31
February
1 2 3 4 5 6
7 8 9 10
\tikz
\calendar [dates=2000-01-20 to 2000-02-10,
week list,month label above right];
• style=month label below left Works like month label above left, only the label is placed below
the row. This placement is not really useful with the week list arrangement, but rather with the
day list right or month list arrangement.
February
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
\tikz
\calendar [dates=2000-02-01 to 2000-02-last,
day list right,month label below left];
• style=month label below centered works like month label above centered, only below.
February
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
\tikz
\calendar [dates=2000-02-01 to 2000-02-last,
day list right,month label below centered];
21.4 Examples
In the following, some example calendars are shown that come either from real applications or are just nice
to look at.
Let us start with a year-2100-countdown, in which we cross out dates as we approach the big celebration.
For this, we set the shape to strike out for these dates.
December 2099
1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31
January 2100
1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31
\begin{tikzpicture}
\calendar
[
dates=2099-12-01 to 2100-01-last,
week list,inner sep=2pt,month label above centered,
month text=\%mt \%y0
]
if (at most=2099-12-29) [nodes={strike out,draw}]
if (weekend) [black!50,nodes={draw=none}]
;
\end{tikzpicture}
The next calendar shows a deadline, which is 10 days in the future from the current date. The last three
days before the deadline are in red, because we really should be done by then. All days on which we can no
longer work on the project are crossed out.
195
18 19 20
21 22 23 24 25 26 27
28 29 30 31
June 2007
1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30
July 2007
1
2 3 4 5 6 7
\begin{tikzpicture}
\calendar
[
dates=\year-\month-\day+-25 to \year-\month-\day+25,
week list,inner sep=2pt,month label above centered,
month text=\textit{\%mt \%y0}
]
if (at least=\year-\month-\day) {}
else [nodes={strike out,draw}]
if (at most=\year-\month-\day+7)
[green!50!black]
if (between=\year-\month-\day+8 and \year-\month-\day+10)
[red]
if (Sunday)
[gray,nodes={draw=none}]
;
\end{tikzpicture}
The following example is a futuristic calendar that is all about circles:
2007
January
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
22
23
24
25
26
27
28
29
30
31
February
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
22
23
24
25
26
27
28
March
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
22
23
24
25
26
27
28
29
30
31
April
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
22
23
24
25
26
27
28
29
30
May
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
22
23
24
25
26
27
28
29
30
31
June
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
22
23
24
25
26
27
28
29
30
July
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
22
23
24
25
26
27
28
29
30
31
August
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
22
23
24
25
26
27
28
29
30
31
September
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
22
23
24
25
26
27
28
29
30
October
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
22
23
24
25
26
27
28
29
30
31
November
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
22
23
24
25
26
27
28
29
30
December
1 2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17 18
19
20
21
22
23
24
25
26
27
28
29
30
31
196
\sffamily
\colorlet{winter}{blue}
\colorlet{spring}{green!60!black}
\colorlet{summer}{orange}
\colorlet{fall}{red}
% A counter, since TikZ is not clever enough (yet) to handle
% arbitrary angle systems.
\newcount\mycount
\begin{tikzpicture}[transform shape]
\tikzstyle{every day}=[anchor=mid,font=\fontsize{6}{6}\selectfont]
\node{\normalsize\the\year};
\foreach \month/\monthcolor in
{1/winter,2/winter,3/spring,4/spring,5/spring,6/summer,
7/summer,8/summer,9/fall,10/fall,11/fall,12/winter}
{
% Computer angle:
\mycount=\month
\advance\mycount by -1
\multiply\mycount by 30
\advance\mycount by -90
% The actual calendar
\calendar at (\the\mycount:6.4cm)
[
dates=\the\year-\month-01 to \the\year-\month-last,
]
if (day of month=1) {\color{\monthcolor}\tikzmonthcode}
if (Sunday) [red]
if (all)
{
% Again, compute angle
\mycount=1
\advance\mycount by -\pgfcalendarcurrentday
\multiply\mycount by 11
\advance\mycount by 90
\pgftransformshift{\pgfpointpolar{\mycount}{1.4cm}}
};
}
\end{tikzpicture}
Next, lets us have a whole year in a tight column:
197
01 1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31 02 1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 03 1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31 04 1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 05 1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31 06 1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 07 1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31 08 1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31 09 1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
10 1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31 11 1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 12 1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31
\begin{tikzpicture}
\small\sffamily
\colorlet{darkgreen}{green!50!black}
\calendar[dates=\year-01-01 to \year-12-31,week list,
month label left,month yshift=0pt,
month text=\textcolor{darkgreen}{\%m0}]
if (Sunday) [black!50];
\end{tikzpicture}
198
22 Entity-Relationship Diagram Drawing Library
\usetikzlibrary{er} % L
A
T
E
X and plain T
E
X
\usetikzlibrary[er] % ConT
E
Xt
This packages provides styles for drawing entity-relationship diagrams.
This library is intended to help you in creating E/R-diagrams. It defines only very little new styles, but
using these style entity instead of saying rectangle,draw makes the code more expressive.
22.1 Entities
The package defines a simple style for drawing entities:
• style=entity This style is to be used with nodes that represent entity types. It causes the node’s
shape to be set to a rectangle that is drawn and whose minimum size and width are set to sensible
values.
Note that this style is called entity despite the fact that it is to be used for nodes representing entity
types (the difference between an entity and an entity type is the same as the difference between an
object and a class in object-oriented programming). If this bothers you, feel free to define a style
entity type instead.
Sheep Genome
\begin{tikzpicture}[node distance=2cm]
\node[entity] (sheep) {Sheep};
\node[entity] (genome) [right of=sheep] {Genome};
\end{tikzpicture}
• style=every entity This stype is envoked by the style entity. To change the appearance of entities,
you can change this style.
Sheep Genome
\begin{tikzpicture}[node distance=2cm]
\tikzstyle{every entity}=[draw=blue!50,fill=blue!20,thick]
\node[entity] (sheep) {Sheep};
\node[entity] (genome) [right of=sheep] {Genome};
\end{tikzpicture}
22.2 Relationships
Relationships are drawn using styles that are very similar to the styles for entities.
• style=relationship This style works like entity, only it is to be used for relationships. Again,
relationships are actually relationship types.
Sheep Genome
has
\begin{tikzpicture}
\node[entity] (sheep) at (0,0) {Sheep};
\node[entity] (genome) at (2,0) {Genome};
\node[relationship] at (1,1.5) {has}
edge (sheep)
edge (genome);
\end{tikzpicture}
• style=every relationship works like every entity
Sheep Genome
has
\begin{tikzpicture}
\tikzstyle{every entity}=[fill=blue!20,draw=blue,thick]
\tikzstyle{every relationship}=[fill=orange!20,draw=orange,thick,aspect=1.5]
\node[entity] (sheep) at (0,0) {Sheep};
\node[entity] (genome) at (2,0) {Genome};
\node[relationship] at (1,1.5) {has}
edge (sheep)
edge (genome);
\end{tikzpicture}
199
22.3 Attributes
• style=attribuate This style is used to indicate that a node is an attribute. To connect an attribute
to its entity, you can use, for example, the child command or the pin option.
Sheep
name
color
\begin{tikzpicture}
\node[entity] (sheep) {Sheep}
child {node[attribute] {name}}
child {node[attribute] {color}};
\end{tikzpicture}
Sheep
name color
\begin{tikzpicture}
\tikzstyle{every pin edge}=[draw]
\node[entity,pin={[attribute]60:name},pin={[attribute]120:color}] {Sheep};
\end{tikzpicture}
• style=key attribute This style is intended for key attributes. By default, the will cause the attribute
to be typeset in italics. Typcially, underlining is used instead, but that looks ugly and it is difficult to
implement in T
E
X.
• style=every attribute This style is used with every (key) attribute.
Sheep
name
Genome
has
\begin{tikzpicture}[text depth=1pt]
\tikzstyle{every attribute}=[fill=black!20,draw=black]
\tikzstyle{every entity}=[fill=blue!20,draw=blue,thick]
\tikzstyle{every relationship}=[fill=orange!20,draw=orange,thick,aspect=1.5]
\node[entity] (sheep) at (0,0) {Sheep}
child {node [key attribute] {name}};
\node[entity] (genome) at (2,0) {Genome};
\node[relationship] at (1,1.5) {has}
edge (sheep)
edge (genome);
\end{tikzpicture}
200
23 Paper Folding Diagrams Library
\usetikzlibrary{folding} % L
A
T
E
X and plain T
E
X
\usetikzlibrary[folding] % ConT
E
Xt
This library defines commands for creating paper folding diagrams. Currently, it just contains a single
command for creating a single diagram, but that one is really useful for creating calendars for your
(real) desktop.
\tikzfoldingdodecahedron[¸options)];
This command draws a folding diagram for a dodecahedron. The syntax is intended to remind of the
\path command, but (currently) you must specify the ¸options) and nothing else may be specified
between the command name and the closing semicolon.
The following ¸options) apply:
• folding line length=¸dimension) sets the length of the base line for folding. For the dodecahe-
dron this is the length of all the sides of the pentagons.
• face 1=¸code) The ¸code) is executed for the first face of the dodecahedron. When it is executed,
the coordinate system will have been shifted and rotated such that it lies at the middle of the first
face of the dodecahedron.
• face 2=¸code) Same as face 1, but for the second face.
• face 3=¸code) Same as face 1, but for the third face.
There are further similar options, ending with the following:
• face 12=¸code) Same as face 1, but for the last face.
Here is a simple example:
1
2
3
4
5
6
7
8
9
1
0
11
1
2
\begin{tikzpicture}[transform shape]
\tikzfoldingdodecahedron
[folding line length=6mm,
face 1={ \node[red] {1};},
face 2={ \node {2};},
face 3={ \node {3};},
face 4={ \node {4};},
face 5={ \node {5};},
face 6={ \node {6};},
face 7={ \node {7};},
face 8={ \node {8};},
face 9={ \node {9};},
face 10={\node {10};},
face 11={\node {11};},
face 12={\node {12};}];
\end{tikzpicture}
The appearance of the cut and folding lines can be influenced using the following styles:
• style=every cut Executed for every line that should be cut using scissors. Empty by default.
• style=every fold Executed for every line that should be folded. Equal to help lines by default.
\begin{tikzpicture}
\tikzstyle{every cut}=[red]
\tikzstyle{every fold}=[dotted]
\tikzfoldingdodecahedron[folding line length=6mm];
\end{tikzpicture}
201
Here is a big example that produces a diagram for a calendar:
January 2007
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31
F
e
b
r
u
a
r
y
2
0
0
7
1
2
3
4
5
6
7
8
9
1
0
1
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
2
6
2
7
2
8
M
a
r
c
h
2
0
0
7
1
2
3
4
5
6
7
8
9
1
0
1
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
2
6
2
7
2
8
2
9
3
0
3
1
A
p
r
i
l
2
0
0
7
1
2
3
4
5
6
7
8
9
1
0
1
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
2
6
2
7
2
8
2
9
3
0
M
a
y
2
0
0
7
1
2
3
4
5
6
7
8
9
1
0
1
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
2
6
2
7
2
8
2
9
3
0
3
1
J u n e 2 0 0 7
1 2 3
4 5 6 7 8 9 1 0
1 1 1 2 1 3 1 4 1 5 1 6 1 7
1 8 1 9 2 0 2 1 2 2 2 3 2 4
2 5 2 6 2 7 2 8 2 9 3 0
J
u
l
y
2
0
0
7
1
2
3
4
5
6
7
8
9
1
0
1
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
2
6
2
7
2
8
2
9
3
0
3
1
A
u
g
u
s
t
2
0
0
7
1
2
3
4
5
6
7
8
9
1
0
1
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
2
6
2
7
2
8
2
9
3
0
3
1
S
e
p
t
e
m
b
e
r
2
0
0
7
1
2
3
4
5
6
7
8
9
1
0
1
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
2
6
2
7
2
8
2
9
3
0
O
c
t
o
b
e
r
2
0
0
7
1
2
3
4
5
6
7
8
9
1
0
1
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
2
6
2
7
2
8
2
9
3
0
3
1
November 2007
1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30
D
e
c
e
m
b
e
r
2
0
0
7
1
2
3
4
5
6
7
8
9
1
0
1
1
1
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
2
0
2
1
2
2
2
3
2
4
2
5
2
6
2
7
2
8
2
9
3
0
3
1
202
\sffamily\scriptsize
\begin{tikzpicture}[transform shape]
\tikzstyle{every calendar}=
[
at={(-8ex,4ex)},
week list,
month label above centered,
month text=\bfseries\textcolor{red}{\%mt} \%y0,
if={(Sunday) [black!50]}
]
\tikzfoldingdodecahedron
[
folding line length=2.5cm,
face 1={ \calendar [dates=\the\year-01-01 to \the\year-01-last];},
face 2={ \calendar [dates=\the\year-02-01 to \the\year-02-last];},
face 3={ \calendar [dates=\the\year-03-01 to \the\year-03-last];},
face 4={ \calendar [dates=\the\year-04-01 to \the\year-04-last];},
face 5={ \calendar [dates=\the\year-05-01 to \the\year-05-last];},
face 6={ \calendar [dates=\the\year-06-01 to \the\year-06-last];},
face 7={ \calendar [dates=\the\year-07-01 to \the\year-07-last];},
face 8={ \calendar [dates=\the\year-08-01 to \the\year-08-last];},
face 9={ \calendar [dates=\the\year-09-01 to \the\year-09-last];},
face 10={\calendar [dates=\the\year-10-01 to \the\year-10-last];},
face 11={\calendar [dates=\the\year-11-01 to \the\year-11-last];},
face 12={\calendar [dates=\the\year-12-01 to \the\year-12-last];}
];
\end{tikzpicture}
203
24 Matrix Library
\usetikzlibrary{matrix} % L
A
T
E
X and plain T
E
X
\usetikzlibrary[matrix] % ConT
E
Xt
This library packages defines additional styles and options for creating matrices.
24.1 Matrices of Nodes
A matrix of nodes is a TikZ matrix in which each cell contains a node. In this case it is bothersome having
to write \node{ at the beginning of each cell and }; at the end of each cell. The following style simplifies
typesetting such matrices.
• style=matrix of nodes Conceptually, this style adds \node{ at the beginning and }; at the end of
each cell and sets the anchor of the node to base. Furthermore, it adds the option name option to each
node, where the name is set to ¸matrix name)-¸row number)-¸column number). For example, if the
matrix has the name my matrix, then the node in the upper left cell will get the name my matrix-1-1.
8 1 6
3 5 7
4 9 2
\begin{tikzpicture}
\matrix (magic) [matrix of nodes]
{
8 & 1 & 6 \\
3 & 5 & 7 \\
4 & 9 & 2 \\
};
\draw[thick,red,->] (magic-1-1) |- (magic-2-3);
\end{tikzpicture}
You may wish to add options to certain nodes in the matrix. This can be achieved in three ways.
1. You can modify, say, the row 2 column 5 option to pass special options to this particular cell.
8 1 6
3 5 7
4 9 2
\begin{tikzpicture}
\tikzstyle{row 2 column 3}=[red]
\matrix [matrix of nodes]
{
8 & 1 & 6 \\
3 & 5 & 7 \\
4 & 9 & 2 \\
};
\end{tikzpicture}
2. At the beginning of a cell, you can use a special syntax. If a cell starts with a vertical bar, then
everything between this bar and the next bar is passed on to the node command.
8 1 6
3 5 7
4 9 2
\begin{tikzpicture}
\matrix [matrix of nodes]
{
8 & 1 & 6 \\
3 & 5 & |[red]| 7 \\
4 & 9 & 2 \\
};
\end{tikzpicture}
You can also use an option like |[red] (seven)| to give a different name to the node.
Note that the & character also takes an optional argument, which is an extra column skip.
8 1 6
3 5 7
4 9 2
\begin{tikzpicture}
\matrix [matrix of nodes]
{
8 &[1cm] 1 &[3mm] |[red]| 6 \\
3 & 5 & |[red]| 7 \\
4 & 9 & 2 \\
};
\end{tikzpicture}
204
3. If your cell starts with a \path command or any command that expands to \path, which includes
\draw, \node, \fill and other, the \node{ startup code and the }; code are suppressed. This
means that for this particular cell you can provide a totally different contents.
8 1 6
3 5 7
4 9 2
\begin{tikzpicture}
\matrix [matrix of nodes]
{
8 & 1 & 6 \\
3 & 5 & \node[red]{7}; \draw(0,0) circle(10pt);\\
4 & 9 & 2 \\
};
\end{tikzpicture}
• style=matrix of math nodes This style is almost the same as the previous style, only $ is added at
the beginning and at the end of each node, so math mode will be switched on in all nodes.
a
8
a
1
a
6
a
3
a
5
a
7
a
4
a
9
a
2
\begin{tikzpicture}
\matrix [matrix of math nodes]
{
a_8 & a_1 & a_6 \\
a_3 & a_5 & |[red]| a_7 \\
a_4 & a_9 & a_2 \\
};
\end{tikzpicture}
• nodes in empty cells=¸true or false) When set to true, a node (with an empty contents) is put in
empty cells. Normally, empty cells are just, well, empty. The style can be used together with both a
matrix of nodes and a matrix of math nodes.
a
8
a
6
a
3
a
7
a
4
a
9
\begin{tikzpicture}
\matrix [matrix of math nodes,nodes={circle,draw}]
{
a_8 & & a_6 \\
a_3 & & a_7 \\
a_4 & a_9 & \\
};
\end{tikzpicture}
a
8
a
6
a
3
a
7
a
4
a
9
\begin{tikzpicture}
\matrix [matrix of math nodes,nodes={circle,draw},nodes in empty cells]
{
a_8 & & a_6 \\
a_3 & & a_7 \\
a_4 & a_9 & \\
};
\end{tikzpicture}
24.2 Delimiters
Delimiters are parantheses or braces to the left and right of a formula or a matrix. The matrix library offers
options for adding such delimiters to a matrix. However, delimiters can actually be added to any node that
has the standard anchors north, south, north west and so on. In particular, you can add delimiters to any
rectangle box. They are implemented by “measuring the height” of the node and then adding a delimiter
of the correct size to the left or right using some after node magic.
• left delimiter=¸delimiter) This option can be given to a any node that has the standard anchors
north, south and so on. The ¸delimiter) can be any delimiter that is acceptable to T
E
X’s \left
command.
a
8
a
1
a
6
a
3
a
5
a
7
a
4
a
9
a
2








\begin{tikzpicture}
\matrix [matrix of math nodes,left delimiter=(,right delimiter=\}]
{
a_8 & a_1 & a_6 \\
a_3 & a_5 & a_7 \\
a_4 & a_9 & a_2 \\
};
\end{tikzpicture}
205

1
0
xdx

\begin{tikzpicture}
\node [fill=red!20,left delimiter=(,right delimiter=\}]
{$\displaystyle\int_0^1 x\,dx$};
\end{tikzpicture}
• style=every delimiter This style is executed for every delimiter. You can use it to shift or color
delimiters or do whatever.
• style=every left delimiter This style is additionally executed for every left delimiter.
a
8
a
1
a
6
a
3
a
5
a
7
a
4
a
9
a
2








\begin{tikzpicture}
\tikzstyle{every left delimiter}=[red,xshift=1ex]
\tikzstyle{every right delimiter}=[xshift=-1ex]
\matrix [matrix of math nodes,left delimiter=(,right delimiter=\}]
{
a_8 & a_1 & a_6 \\
a_3 & a_5 & a_7 \\
a_4 & a_9 & a_2 \\
};
\end{tikzpicture}
• right delimiter=¸delimiter) Works as above.
• style=every right delimiter Works as above.
• above delimiter=¸delimiter) This option allows you to add a delimiter above the node. It is imple-
menting by rotating a left delimiter.
a
8
a
1
a
6
a
3
a
5
a
7
a
4
a
9
a
2











    
        
\begin{tikzpicture}
\matrix [matrix of math nodes,%
left delimiter=\|,right delimiter=\rmoustache,%
above delimiter=(,below delimiter=\}]
{
a_8 & a_1 & a_6 \\
a_3 & a_5 & a_7 \\
a_4 & a_9 & a_2 \\
};
\end{tikzpicture}
• style=every above delimiter Works as above.
• below delimiter=¸delimiter) works as above.
• style=every below delimiter Works as above.
206
25 Mindmap Drawing Library
\usetikzlibrary{mindmap} % L
A
T
E
X and plain T
E
X
\usetikzlibrary[mindmap] % ConT
E
Xt
This packages provides styles for drawing mindmap diagrams.
25.1 Overview
This library is intended to make the creation of mindmaps easier. A mindmap is a graphical representation
of a concept together with related concepts and annotations. Mindmaps are, essentially, trees, possibly with
a few extra edges added, but they are usually drawn in a special way: The root concept is placed in the
middle of the page and is drawn as a huge circle, ellipse, or cloud. The related concepts then “leave” this
root concept in branch-like tendrils.
The mindmap library of TikZ produces mindmaps that look a bit different from the standard mindmaps:
While the big root concept is still a circle, related concepts are also depicted as (smaller) circles. The related
concepts are linked to the root concept via organic-looking connections. The overall effect is visually rather
pleasing, but readers may not immediately think of a mindmap when they see a picture created with this
library.
Although it is not strictly necessary, you will usually create mindmaps using TikZ’s tree mechanism
and some of the styles and macros of the package work best when used inside trees. However, it is still
possible and sometimes necessary to treat parts of a mindmap as a graph with arbitrary edges and this is
also possible.
25.2 The Mindmap Style
Every mindmap should be put in a scope or a picture where the mindmap style is used. This style installs
some internal settings.
• style=mindmap Use this style with all pictures or at least scopes that contain a mindmap. It installs
a whole bunch of settings that are useful for drawing mindmaps.
Root concept Child concept
\tikz[mindmap,concept color=red!50]
\node [concept] {Root concept}
child[grow=right] {node[concept] {Child concept}};
The sizes of concepts are predefined in such a way that a medium-size mindmap will fit on an A4 page
(more or less).
• style=every mindmap This style is included by the mindmap style. Change this style to add special
settings to your mindmaps.
207
Root concept
Child concept
\tikz[large mindmap,concept color=red!50]
\node [concept] {Root concept}
child[grow=right] {node[concept] {Child concept}};
• style=large mindmap This style includes the mindmap style, but additionally changes the default size
of concepts and of distances so that a medium-sized mindmap will fit on an A3 page (A3 pages are
twice as large as A4 pages).
• style=huge mindmap This style causes conepts to be even bigger and it is best used with A2 paper
and above.
25.3 Concepts Nodes
The basic entities of mindmaps are called concepts in TikZ. A concept is a node of style concept and it
must be circular for some of the connection macros to work.
25.3.1 Isolated Concepts
The following styles influence how isolated concepts are rendered:
• style=concept This style should be used with all nodes that are concepts, although some styles like
extra concept install this style automatically.
Bascially, this style makes the concept node circular and installs a uniform color called concept color,
see below. Additionally, the style every concept is called.
Some concept
\tikz[mindmap,concept color=red!50] \node [concept] {Some concept};
• style=every concept In order to change the appearance of concept nodes, you should change this
style. Note, however, that the color of a concept should be uniform for some of the connection bar
stuff to work, so you should not change the color or the draw/fill state of concepts using this option.
It is mostly useful for changing the text color and font.
• concept color=¸color) This option tells TikZ which color should be used for filling and stroking
concepts. The difference between this option and just setting every concept to the desired color is
that this option allows TikZ to keep track of the colors used for concepts. This is important when
208
you change the color between two connected concepts. In this case, TikZ can automatically create a
shading that provides a smooth transition between the old and the new concept color; we will come
back to this in the next section.
• style=extra concept This style is intended for concepts that are not part of the “mindmap tree,”
but stand beside it. Typically, they will have a subdued color are be smaller. In order to have these
concepts appear in a uniform way and in order to indicate in the code that these concepts are extra,
you can use this style.
Root concept
extra
concept
\begin{tikzpicture}[mindmap,concept color=blue!80]
\node [concept] {Root concept};
\node [extra concept] at (10,0) {extra concept};
\end{tikzpicture}
• style=every extra concept Change this style to change the appearance of extra concepts.
25.3.2 Concepts in Trees
As pointed out earlier, TikZ assumes that your mindmap is build using the child facilities of TikZ. There
are numerous options that influence how concepts are rendered at the different levels of a tree.
• style=root concept This style is used for the roots of mindmap trees. More precisely, this style is
included by the mindmap style. Thus, by adding something to this, you can change how the root of a
mindmap will be rendered.
Root concept
\tikzstyle{root concept}+=[concept color=blue!80,minimum size=3.5cm]
\tikz[mindmap] \node [concept] {Root concept};
Note that styles like large mindmap redefine these styles, so you should add something to this style
only inside the picture.
• style=level 1 concept The mindmap style adds this style to the level 1 style. This means that the
first level children of a mindmap tree will use this style.
209
Root concept
child
child
\tikzstyle{root concept}+=[concept color=blue!80]
\tikzstyle{level 1 concept}+=[concept color=red!50]
\tikz[mindmap]
\node [concept] {Root concept}
child[grow=30] {node[concept] {child}}
child[grow=0 ] {node[concept] {child}};
• style=level 2 concept works like level 1 concept, only for second level children.
• style=level 3 concept works like level 1 concept.
• style=level 4 concept works like level 1 concept. Note that there are not fifth and higher level
styles, you need to modify level 5 directly in such cases.
• concept color=¸color) We saw already that this option is used to change the color of concepts. We
now have a look at its effect when used on child nodes of a concept. Normally, this option simply
changes the color of the children. However, when the option is given as an option to the child
operation (and not to the node operation and also not as an option to all children via the level 1
style), TikZ will smoothly change the concept color from the parent’s color to the color of the child
concept.
Here is an example:
Root concept
Child concept
Child concept
\tikz[mindmap,concept color=blue!80]
\node [concept] {Root concept}
child[concept color=red,grow=30] {node[concept] {Child concept}}
child[concept color=orange,grow=0] {node[concept] {Child concept}};
In order to have all children of a certain level have a different concept color, a tiny bit of magic is
needed:
210
Root concept
child
child
\tikzstyle{root concept}+=[concept color=blue]
\tikzstyle{level 1 concept}+=[set style={{every child}=[concept color=blue!50]}]
\tikz[mindmap,text=white]
\node [concept] {Root concept}
child[grow=30] {node[concept] {child}}
child[grow=0 ] {node[concept] {child}};
25.4 Connecting Concepts
25.4.1 Simple Connections
The easiest way to connect two concepts is to draw a line between them. In order to give such lines a
consistent appearance, it is recommendable to use the following style when drawing such lines:
• style=concept connection This style can be used for lines between two concepts. Feel free to redefine
this style.
A problem arises when you need to connect concepts after the main mindmap has been drawn. In this
case you will want the connection lines to lie behind the main mindmap. However, you can draw the lines
only after the coordinates of the concepts have been determined. In this case you should place the connecting
lines on a background layer as in the following example:
Root concept
child
child
child
211
\tikzstyle{root concept}+=[concept color=blue!20,minimum size=2cm]
\tikzstyle{level 1 concept}+=[sibling angle=45]
\begin{tikzpicture}[mindmap]
\node [concept] {Root concept}
[clockwise from=45]
child { node[concept] (c1) {child}}
child { node[concept] (c2) {child}}
child { node[concept] (c3) {child}};
\begin{pgfonlayer}{background}
\draw [concept connection] (c1) edge (c2)
edge (c3)
(c2) edge (c3);
\end{pgfonlayer}
\end{tikzpicture}
25.4.2 The Circle Connection Bar Snake
Instead of a simple line between two concepts, you can also add a bar between the two nodes that has slightly
organic ends. These bars are also used by default as the edges from parents in the mindmap tree.
For the drawing of the bars a special snake is used, which is defined in the mindmap library:
Snake circle connection bar
This snake can be used to connect two circles. The start of the snake should lie on the border of the
first circle, the end should lie on the border of the second circle. The following two macros should be
initialized with the sizes of the circles:
• \pgfsnakecirclestartradius
• \pgfsnakecircleendradius
Furthermore, the following two macros influence the snake:
• \pgfsnakesegmentamplitude
• \pgfsnakesegmentangle
The snake itself will be a path that starts on the border of the first circle at the specified angle relative
to the line connecting the centers of the circles. The path then changes into a rectangle whose thickness
is given by the amplitude. Finally, the path ends with the same angles on the second circle.
Here is an example that should make this clearer:
\begin{tikzpicture}
\fill[blue!20] (0,0) circle (1cm);
\fill[red!20] (2.5,0) circle (.5cm);
\def\pgfsnakecirclestartradius{1cm}
\def\pgfsnakecircleendradius{.5cm}
\def\pgfsnakesegmentamplitude{2mm}
\def\pgfsnakesegmentangle{30}
\filldraw [draw=red,fill=black,snake=circle connection bar] (1,0) -- (2,0);
\end{tikzpicture}
As can be seen, the snake consists of three parts and is not really useful for drawing. However, if you
fill the snake only and if you use the same color as for the circles, the result is better.
\begin{tikzpicture}[blue!50]
\fill (0,0) circle (1cm);
\fill (2.5,0) circle (.5cm);
\def\pgfsnakecirclestartradius{1cm}
\def\pgfsnakecircleendradius{.5cm}
\def\pgfsnakesegmentamplitude{2mm}
\def\pgfsnakesegmentangle{30}
\fill [snake=circle connection bar] (1,0) -- (2,0);
\end{tikzpicture}
In the above example you may notice the small white line between the circles and the snake. This is
due to rounding errors. Unfortunately, for larger distances, there errors can accumulate quite strongly,
212
especially since TikZ and T
E
X are not very good at computing square roots. For this reason, it is a good
idea to make the circles slightly larger to cover up such problems. When using nodes of shape circle,
you can just add the draw option with a line width or one or two points (for very large distances you
may need line width up to 4pt).
\begin{tikzpicture}[blue!50]
\fill (0,0) circle (1cm+1pt);
\fill (2.4,0) circle (.5cm+1pt);
\def\pgfsnakecirclestartradius{1cm}
\def\pgfsnakecircleendradius{.5cm}
\def\pgfsnakesegmentamplitude{2mm}
\def\pgfsnakesegmentangle{30}
\fill [snake=circle connection bar] (1,0) -- (1.9,0);
\end{tikzpicture}
Note the slightly strange outer sep=0pt. This is needed so that the snakes being on the border of the
filled circle, not on the border of the stroked circle (which is slightly larger and this slightly larger size
is exactly what we wish to use to cover up the rounding errors).
25.4.3 The Circle Connection Bar To-Path
The circle connection bar snake is a bit complicated to use. Especially specifying the radii is quite bothersome
(the amplitude and the angle can be set once and for all). For this reason, the mindmap library defines a
special to-path, that performs the necessary computations for you.
• style=circle connection bar This style installs a rather involved to-path. Unlike normal to-paths,
this path requires that the start and the target of the to-path are named nodes of shape circle – if
this is not the case, this path will produce errors.
Assuming that the start and the target are circles, the to-path will first compute the radii of these
circles (by measuring the distance from the center anchor to some anchor on the border) and will
set the \pgfsnakecirlce... macros accordingly. Next, the fill option is set to the concept color
while draw=none is set. The snake is set to circle conncetion bar. Finally, the following style is
included:
– style=every circle connection bar Redefine this sytle to change the appearance of circle
connection bar to-paths.
\begin{tikzpicture}[concept color=blue!50,blue!50,outer sep=0pt]
\node (n1) at (0,0) [circle,minimum size=2cm,fill,draw,thick] {};
\node (n2) at (2.5,0) [circle,minimum size=1cm,fill,draw,thick] {};
\path (n1) to[circle connection bar] (n2);
\end{tikzpicture}
Note that it is not a good idea to have more than one to operation together this the option
circle connection bar in a single \path. Use the edge operation, instead, for creating multiple
connections and this operation creates a new scope for each edge.
In a mindmap we sometimes want colors to change from one concept color to another. Then, the
connection bar should, ideally, consist of a smooth transition between these two colors. Getting this right
using shadings is a bit tricky if you try this “by hand,” so the mindmap library provides a special option for
facilitating this procedure.
• circle connection bar switch color=from (¸first color)) to (¸second color)) This style works
similarly to the circle connection bar. The only difference is that instead of filling the path with
a single color a shading is used.
213
\begin{tikzpicture}[outer sep=0pt]
\node (n1) at (0,0) [circle,minimum size=2cm,fill,draw,thick,red] {};
\node (n2) at (30:2.5) [circle,minimum size=1cm,fill,draw,thick,blue] {};
\path (n1) to[circle connection bar switch color=from (red) to (blue)] (n2);
\end{tikzpicture}
25.4.4 Tree Edges
Most of the time, concepts in a mindmap are connected automatically when the mindmap is build as a tree.
The reason is that the mindmap installs a circle connection bar path as the edge from parent path. Also,
the mindmap option takes care of things like setting the correct draw and outer sep settings and some other
stuff.
In detail, the mindmap option sets the edge from parent path to a path that uses the to-path
circle connection bar to connect the parent node and the child node. The concept color option (lo-
cally) changes this by using circle connection bar switch color instead with the from-color set to the
old (parent’s) concept color and the to-color set to the new (child’s) concept color. This menas that when
you provide the concept color option to a child command, the color will change from the parent’s concept
color to the specified color.
Here is an example of a tree build in this way:
Computer Science practical
algorithms
data
structures
pro-
gramming
languages
software
engineering
applied
databases
WWW
technical
theoretical
214
\begin{tikzpicture}
\path[mindmap,concept color=black,text=white]
node[concept] {Computer Science}
[clockwise from=0]
child[concept color=green!50!black] {
node[concept] {practical}
[clockwise from=90]
child { node[concept] {algorithms} }
child { node[concept] {data structures} }
child { node[concept] {pro\-gramming languages} }
child { node[concept] {software engineer\-ing} }
}
child[concept color=blue] {
node[concept] {applied}
[clockwise from=-30]
child { node[concept] {databases} }
child { node[concept] {WWW} }
}
child[concept color=red] { node[concept] {technical} }
child[concept color=orange] { node[concept] {theoretical} };
\end{tikzpicture}
25.5 Adding Annotations
An annotation is some text outside a mindmap that, unlike an extra concept, simply explains something
in the mindmap. The following style is mainly intended to help readers of the code see that a node in an
annotation node.
• style=annotation This style indicates that a node is an annotation node. It includes the style
every annotation, which allows you to change this style in a convenient fashion.
Root concept
The root concept is, in general,
the most important concept.
\tikzstyle{every annotation}=[fill=red!20]
\begin{tikzpicture}[mindmap,concept color=blue!80]
\node [concept] (root) {Root concept};
\node [annotation,right] at (root.east)
{The root concept is, in general, the most important concept.};
\end{tikzpicture}
• style=every annotation This style is included by annotation.
25.6 Examples
The following pictures show examples of more complicated mindmaps that have been created using the
mindmap library. Note that these mindmaps should be printed on A1 paper.
215
INFORMATIK
TECHNISCHE
INFORMATIK
Prozessoren Robotik
Schaltkreise
Peripherie Maus
Bildschirm
Drucker
Tastatur
Netzwerkkarte
Netzwerke
Internet
Funk
Bluetooth
Wirless LAN
Kabel Telefon
ISDN
DSL
THEORETISCHE
INFORMATIK
Typtheorie
Maschinenmodelle
K¨unstliche
Intelligenz
Automatentheorie
NFAs
DFAs
Theorie
formaler Sprachen
Spachen
Grammatiken
regul¨ ar
kontextfrei
regul¨ are
Ausdr¨ucke
Komplexit¨ ats-
theorie
Ressourcen
Klassen
PRAKTISCHE
INFORMATIK
Algorithmen
Graph-
algorithmen
geometrische
algorithmische
Methoden
Dynamische
Programme
Rekursion
Greedy-
Heuristiken
Back-
tracking
Approxi-
mation
Sortieren
Suchen
Optimierung
Datenstrukturen
zusammen-
gesetzte
Graphen
B¨ aume
Hashing
Arrays
Listen
einfache
Strings
Zahlen
Programmier-
sprachen
imperativ
C
Assembler
Pascal
objektorientiert
Java
C++
Smalltalk
funktional
ML
Scheme
logisch
Prolog
Softwaretechnik
Modellierung
Strukturierung
Modulari-
sierung
Planung
ANGEWANDTE
INFORMATIK
Datenbanken
Theorie
ER-
Diagramme
Relationen-
Algebren
Datenbank-
Sprachen
SQL
Datenbank-
Systeme
DB2
MySQL
Anwendung
Mail
SPSS
Matlab
Word
Betriebssysteme
Windows
Unix
Shells
WWW
ELEKTROTECHNIK
MASCHINENBAU
LINUGISTIK
MATHEMATIK
WIRTSCHAFT
Vorlesungskarte zu den Vorlesungen
Informatik A f¨ur Molecular Life Sciences 2006
Informatik B f¨ur Molecular Life Sciences 2007
Veranstaltungsziele
– Informationsverarbeitende Systeme theoretisch verstehen
– Informationsverarbeitende Systeme in Forschungs- und
Entwicklungsprojekten benutzen k¨ onnen
– Algorithmen und Datenstrukturen verschiedenen Projektbed¨urfnissen
anpassen k¨ onnen
– Vorbereitung auf vertiefende Informatikveranstaltungen
D

i=1
1
n2 =
π2
6
Satz
Subjekt
Das Sein
ist.
A1 Information und Daten
– Die bin¨ are Speicherform verschiedener Daten
kennen und verstehen
– Wichtige Einheiten von Information kennen und
die Speichergr¨ oße von Daten absch¨ atzen
k¨ onnen
– Beliebige Daten kodieren k¨ onnen
– Das Konzept der Datenkompression kennen und
verstehen
A2 Computer-Hardware
– Grundlegende Rechnerarchitekturen kennen
– Klassen von Computern kennen und ihre
Leistungsf¨ ahigkeit absch¨ atzen k¨ onnen
– Den prinzipiellen Aufbau von Computern
beschreiben k¨ onnen
– Die wichtigsten Hardwarekomponenten kennen
und ihre Leistungsf¨ ahigkeit absch¨ atzen k¨ onnen
A3 Computer-Software
– Ein-Prozess-Systeme und das EVA-Prinzip
verstehen
– Mehr-Prozess-Systeme verstehen
– Programme als Instruktionsfolgen begreifen
– Verschiedene Arten von Software benennen und
unterscheiden k¨ onnen
A5 Shells
– Unix- und Programm-Shells benutzen k¨ onnen
– Wichtige Unix-Kommandos kennen
– Einfache Shell-Skripte erstellen k¨ onnen
A4 Betriebssysteme
– Aufgaben von Betriebssystemen kennen
– Konzepte der Benutzer-, Prozess-, und
Dateiverwaltung kennen
– Mit Datein und Dateirechten umgehen k¨ onnen
– Begriff der Shell kennen
A6 Der Algorithmusbegriff
– Den Begriff des Algorithmus verstehen
– Einfache L¨ osungsideen als Algorithmen
formulieren k¨ onnen
– Probleme spezifizieren k¨ onnen
– Programmiersprachen als Kommunikationsmittel
f¨ur Algorithmen begreifen
– Das Konzept des ¨ Ubersetzers verstehen
A7 Imperative Programmierung
– Zuweisungen verstehen und benutzen k¨ onnen
– Grundlegende Kontrollstrukturen verstehen und
benutzen k¨ onnen
– Algorithmen als imperative Java-Programme
formulieren k¨ onnen
A8 Die Programmiersprache Java
– Syntax des imperativen Kerns von Java
beherrschen
– Java-Programme selbstst¨ andig erstellen k¨ onnen
– Java-Programme ¨ubersetzen und testen k¨ onnen
A9 Elementare Datentypen
– Variablendeklarationen verstehen und benutzen
k¨ onnen
– Elementare Datentypen von Java kennen
– Typkorrektheit und Typefehler verstehen
A10 Zeichenketten
– Zeichenketten (Strings) als Datentypen kennen
– Einfache Stringalgorithmen kennen und
implementieren k¨ onnen
– Fortgeschrittene Stringalgorithmen kennen und
implementieren k¨ onnen
A11 Arrays (Felder)
– Konzept des Arrays verstehen
– Java-Syntax von Arrays beherrschen
– Java-Programme mit Array-Nutzung
implementieren k¨ onnen
A12 Scoping
– Scoping verstehen und anwenden k¨ onnen
– Die Speicherung von Variablen verstehen
A13 Modularisierung im Kleinen
– Das Konzept des Unterprogramms/Methode
verstehen und anwenden k¨ onnen
– Java-Syntax der Modularisierung auf
Methodenebene beherrschen
A14 Suchalgorithmen
– Lineare Suche verstehen und implementieren
k¨ onnen
– Bin¨ are Suche verstehen und implementieren
k¨ onnen
– Laufzeit der Verfahren vergleichen k¨ onnen
– Absch¨ atzen k¨ onnen, wann sich Vorsortierung
lohnt
A15 Sortieralgorithmen
– Arten von Sortierproblemen kennen
– Bubble-Sort, Insertion-Sort, Selection-Sort
verstehen und implementieren k¨ onnen
A16 Laufzeit von Algorithmen
– Wichtige Ressourcemaße verstehen und erkl¨ aren
k¨ onnen
– Das Laufzeitverhalten der wichtigsten
Algorithmen kennen
– Das Laufzeitverhalten einfacher Programme
absch¨ atzen k¨ onnen
– Die Gruppierung von Problemen nach groben
Laufzeitklassen verstehen
A17 Einf¨uhrung zur Rekursion
– Das Konzept der Rekursion verstehen
– Rekursive Methoden implementieren k¨ onnen f¨ur
einfache Probleme
A18 Rekursion und schnelles Sortieren
– Merge-Sort verstehen und implementieren
k¨ onnen
– Quick-Sort verstehen und implementieren k¨ onnen
A19 Modularisierung im Großen
– Das Konzept des Top-Down-Entwurf verstehen
und anwenden k¨ onnen
– Das Konzept der Klassen- und Modulbildung
verstehen
A20 Klassen und Modellierung
– Konzept der Klasse verstehen
– Modellierung als Methodik kennen
– Statische Klassenmodelle von
Anwendungsszenarien erstellen k¨ onnen
A21 Klassen und Objekte
– Lebenszyklus von Objekten kennen
– Neue Datentypen deklarieren und benutzen
k¨ onnen
– Nichtstatische Methoden kennen und benutzen
k¨ onnen
– Java-Syntax f¨ur Objekte kennen
A22 Listen als Datenstruktur
– Das Konzept der komplexen Datenstruktur
verstehen
– Die Liste als Datenstruktur verstehen und in Java
implementieren k¨ onnen
– Den Unterschied zwischen Listen und Arrays
erkl¨ aren k¨ onnen
A23 Angewandte Listen
– Fortgeschrittene Arten von Listen kennen
– Vor- und Nachteile von Listen und Arrays
benennen k¨ onnen
– Das Konzept der abstrakten Datenstuktur kennen
A24 B¨ aume
– Das Konzept des (Such)Baumes verstehen
– (Such)B¨ aume in Java implementieren k¨ onnen
– Baumtraversierung kennen und implementieren
k¨ onnen
A25 Suchb¨ aume
– Suchb¨ aume in Java implementieren k¨ onnen
– 2-3-Suchb¨ aumen kennen und verstehen
– Vor- und Nachteile von Arrays, Listen, einfachen
B¨ aumen und fortgeschrittenen Suchb¨ aumen
beurteilen k¨ onnen
A26 Graphen
– Graphen als mathematische Objekte verstehen
– Daten und ihre Struktur mittels Graphen
modellieren k¨ onnen
– Die Datenstruktur des Graphen verstehen und
implementieren k¨ onnen
A27 Graphalgorithmen
– Einige einfache und schwierige Graphprobleme
kennen
– Arten der Graphtraversierung kennen
– Einen K¨urzeste-Wege-Algorithmus kennen
B1 Datenbanken
– Das Konzept der Datenbank kennen und
verstehen
– Das relationale Datenbankmodell kennen
– Daten mittels des relationalen Modells
modellieren k¨ onnen
B2 Relationenalgebren
– Konzept der Relation-Algebra kennen
– Anfragen mittels Operationen formulieren
k¨ onnen
– Joins verstehen und anwenden k¨ onnen
B3 SQL
– Syntax von SQL beherrschen
– SQL-Queries erstellen k¨ onnen
– Mit Datenbanken mittels einer SQL-Shell
kommunizieren k¨ onnen
B4 Das World-Wide-Web
– Aufbau des WWW kennen
– Aufbau von HTML kennen
– Unterschied zwischen Web-Clients und
Web-Servern kennen
– Eigene Webseiten erstellen k¨ onnen
B5 Rechnernetze und das Internet
– Konzept des Kommunikationsprotokolls kennen.
– OSI-Schichtenmodell kennen.
– Unterschied zwischen LAN und Internet kennen.
– Vor- und Nachteile netzbasierter Projekte und
Ans¨ atze beurteilen k¨ onnen.
B6 IT-Sicherheit
– Bedrohungsszenarien der Sicherheit von
IT-Systemen kennen und einsch¨ atzen k¨ onnen
– Schutzmaßnahmen f¨ur die Sicherheit von
IT-Systemen kennen und ergreifen k¨ onnen
B7 Formale Grammatiken
– Das Konzept der formalen Sprache verstehen
– Probleme als formale Sprachen formulieren
k¨ onnen
– Syntax von kontextfreien und regul¨ aren
Grammatiken kennen
– Ableitungen bilden k¨ onnen und die erzeugte
Sprache angeben k¨ onnen
– Probleme mittels formalen Grammatiken
beschreiben k¨ onnen
B8 Endliche Automaten
– Funktionsweise von endlichen Automaten
verstehen
– Formale Sprachen mittels endlicher Automaten
entscheiden k¨ onnen
– Endliche Automaten implementieren k¨ onnen
– Einsch¨ atzen k¨ onnen, ob eine Sprache regul¨ ar ist
B9 Regul¨ are Ausdr¨ucke
– Konzept des regul¨ aren Ausdrucks verstehen
– Probleme als regul¨ are Ausdr¨ucke beschreiben
k¨ onnen
– Verh¨ altnis von regul¨ aren Ausdr¨ucken und
endlichen Automaten kennen
B10 Optimierungsprobleme
– Probleme als Entscheidungsprobleme,
Funktionsprobleme und Optimierungsprobleme
formulieren k¨ onnen
– Die Konzepte Optimierungsproblem, L¨ osung und
Maße verstehen
– Schwierige und leichte Optimierungsprobleme
kennen
B11 Greedy-Heuristiken
– Das Konzept der heuristischen L¨ osung verstehen
– Die Idee von Greedy-Heuristiken kennen
– Beispiele von Greedy-Heuristiken kennen
– Greedy-Heuristiken entwerfen k¨ onnen
B12 Backtracking-Algorithmen
– Die Idee des Backtracking verstehen
– Backtracking-Algorithmen implementieren
k¨ onnen
B13 Approximationsalgorithmen
– Das Konzept der approximativen L¨ osung
verstehen
– L¨ osungsverfahren f¨ur konkrete Probleme kennen
O
k
to
b
e
r
2
0
0
6
N
o
v
e
m
b
e
r
2
0
0
6
D
e
z
e
m
b
e
r
2
0
0
6
J
a
n
u
a
r
2
0
0
7
F
e
b
r
u
a
r
2
0
0
7
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
23
24
25
26
27
30
31
21
22
28
29
1
2
3
6
7
8
9
10
13
14
15
16
17
20
21
22
23
24
27
28
29
30
4
5
11
12
18
19
25
26
1
4
5
6
7
8
11
12
13
14
15
18
19
20
21
22
2
3
9
10
16
17
23
24
25
26
27
28
29
30
31
8
9
10
11
12
15
16
17
18
19
22
23
24
25
26
29
30
31
13
14
20
21
27
28
1
2
3
4
5
6
7
1
2
5
6
7
8
9
3
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Informatik-Einf¨uhrung
Java-Einf¨uhrung
Algorithmik
Modellierung
Datenstrukturen
Einf¨uhrung, Erwartungsabfrage
Information und Daten
Computer-Hardware
Computer-Software
Betriebssyteme
Shells
Der Algorithmus-Begriff
Imperative Programmierung
Programmiersprache Java
Elementare Datentypen
Zeichenketten
Arrays
Scoping
Modularisierung im Kleinen
Suchalgorithmen
Sortieralgorithmen
Laufzeit von Algorithmen
Einf¨uhrung Rekursion
Rekursives Sortieren
Modularisierung im Großen
Klassen und Modellierung
Klassen und Objekte
Listen I
Listen II
B¨ aume I
B¨ aume II
Graphen I
Graphen II
Zusammenfassung, Evaluation
A
p
r
il
2
0
0
7
M
a
i
2
0
0
7
J
u
n
i
2
0
0
7
J
u
li
2
0
0
7
1
2
3
4
5
6
9
10
11
12
13
16
17
18
19
20
23
24
25
26
27
30
7
8
14
15
21
22
28
29
1
2
3
4
7
8
9
10
11
14
15
16
17
18
21
22
23
24
25
28
29
30
31
5
6
12
13
19
20
26
27
1
4
5
6
7
8
11
12
13
14
15
18
19
20
21
22
25
26
27
28
29
2
3
9
10
16
17
23
24
30
2
3
4
5
6
9
10
11
12
13
16
17
18
19
20
1
7
8
14
15
21
22
23
24
25
26
27
28
29
30
31
Datenbanken
Internet und WWW
Theorie der Zeichenketten
Theorie schwieriger Probleme
Einf¨uhrung
Datenbanken
Relationen-Algebren
SQL
WWW
Internet
IT-Sicherheit
Grammatiken
Endliche Automaten
Regul¨ are Ausdr¨ucke
Optimierung
Greedy-Heuristiken
Backtracking
Approximation
Zusammenfassung, Evaluation
Stand vom 8. Oktober 2006
www.tcs.uni-luebeck.de/pages/tantau/teaching.html – Lehre von Till Tantau
UN
I V E R S I T ¨ A T ZU L ¨ UB E CK
I N
S
T
I T
U
T
F ¨ U R
T H E OR E T I S C H E
I N
F O
R
M
A
T
I K
Theoretical
Computer
Science
1
LOGIK
BEWEISE Kalk¨ule
Hilbertkalk¨ul
Resolutions-
kalk¨ul
Sequenzen-
kalk¨ule
nat¨urliches
Schließen
Axiomensysteme Arten
Zermelo-
Fraenkel
Neumann-
Bernay-
G¨ odel
Frege
Eigenschaften
Vollst¨ andig-
keit
Korrektheit
Ableitungen Modus ponens
Substitution
Erstellung
von Menschen
f¨ur Menschen
von Menschen
f¨ur Maschinen
von Maschinen
f¨ur Maschinen
LOGISCHE
SYSTEME
Aussagenlogik
Pr¨ adikatenlogik
erster Stufe
zweiter Stufe
higher-
order logic
Intuitionistische
Logik
Gleichungslogik
Temporallogik
Modallogik
SYNTAX
Variablen
Aussagen-
variablen
Individuen-
variablen
Normalformen
disjunktive
Normalform
konjunktive
Normalform
Signaturen
Formeln
Junktoren
Quantoren
freie und
gebundene
Variablen
Terme
Konstanten-
symbole
Funktions-
symbole
Klammerung
Formale Sprachen
Grammatiken
Sprachen
Alphabete
und Strings
SEMANTIK
Modelle Belegung
Auswertung
Modellrelation
Theorie
logische
Strukturen
Universum
Konstanten
Funktionen
Relationen
G¨ultigkeit Wahrheitstafel
Belegung
logische
Folgerung
Implikation
¨ Aquivalenz
Tautologie
Kontradiktion
K¨ UNSTLICHE
INTELLIGENZ
THEORETISCHE
INFORMATIK
FUNKTIONALE
PROGRAM-
MIERUNG
MENGENLEHRE
DATENBANKEN
TECHNISCHE
INFORMATIK
MODEL-CHECKING
IT-SICHERHEIT
Vorlesungskarte zur Vorlesung
Logik f¨ur Informatiker 2006
Veranstaltungsziele
– Verst¨ andnis der Konzepte Syntax und Semantik an den Beispielen
Aussagen- und Pr¨ adikatenlogik erwerben
– Formale Systeme und Beweissysteme anwenden k¨ onnen
– Methoden der Logik in praktischen Anwendungen einsetzen k¨ onnen
– Diskrete Problemstellungen formalisieren k¨ onnen
O
k
to
b
e
r
2
0
0
6
N
o
v
e
m
b
e
r
2
0
0
6
D
e
z
e
m
b
e
r
2
0
0
6
J
a
n
u
a
r
2
0
0
7
F
e
b
r
u
a
r
2
0
0
7
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
23
24
25
26
27
30
31
21
22
28
29
1
2
3
6
7
8
9
10
13
14
15
16
17
20
21
22
23
24
27
28
29
30
4
5
11
12
18
19
25
26
1
4
5
6
7
8
11
12
13
14
15
18
19
20
21
22
2
3
9
10
16
17
23
24
25
26
27
28
29
30
31
8
9
10
11
12
15
16
17
18
19
22
23
24
25
26
29
30
31
13
14
20
21
27
28
1
2
3
4
5
6
7
1
2
5
6
7
8
9
3
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Einf¨uhrung
Aussagenlogik
Modellierung
Pr¨ adikatenlogik
Einf¨uhrung, Erwartungsabfrage
Syntax versus Semantik
Grammatiken
Syntax
Semantik
Folgerungsbegriff
Kalk¨ule
Vollst¨ andigkeit
E/R-Modelle
Syntax I
Semantik I
Syntax und Semantik II
Kalk¨ule
Zusammenfassung, Evaluation
1 Syntax versus Semantik
– Die Begriffe Syntax und Semantik erkl¨ aren
k¨ onnen
– Syntaktische und semantische Elemente
nat¨urlicher Sprachen und von
Programmiersprachen benennen k¨ onnen
– Die Begriffe Alphabet und Wort kennen
– Objekte als Worte kodieren k¨ onnen
2 Grammatiken
– Konzepte der formalen Sprache und der
formalen Grammatiken kennen
– Konzepte der Regel und der Ableitung
anwenden k¨ onnen
– Syntax mittels formaler Grammatiken
beschreiben k¨ onnen
3 Aussagenlogik: Syntax
– Konzepte der formalen Variable und formalen
Formel verstehen
– Die wichtigsten Junktoren kennen
– Sicherheitstechnische nat¨urlichsprachliche
Problemstellungen aussagenlogisch formalisieren
k¨ onnen
4 Aussagenlogik: Semantik
– Konzept der Variablenbelegung kennen und
anwenden k¨ onnen
– Die Begriffe der Erf¨ullbarkeit, Tautologie und
Kontradiktion kennen
– Aussagenlogische Formeln auf Eigenschaften
wie Erf¨ullbarkeit ¨uberpr¨ufen k¨ onnen
– Eigenschaften sicherheitstechnischer Systeme
belegen k¨ onnen
5 Aussagenlogik: Folgerung
– Die Begriffe der logischen Folgerung und
¨ Aquivalenz kennen
– Formeln auf ¨ Aquivalenz und
Folgerungsbeziehungen untersuchen k¨ onnen
– Normalformen kennen
– Formeln in Normalformen umwandeln k¨ onnen
6 Aussagenlogik: Kalk¨ule
– Unterschied zwischen formalen und normalen
Beweisen kennen
– Resolutionskalk¨ul f¨ur die Aussagenlogik kennen
– Hilbertkalk¨ul f¨ur die Aussagenlogik kennen und
anwenden k¨ onnen
7 Aussagenlogik: Vollst¨ andigkeit
– Das Axiomensystem von Frege kennen
– Die Begriffe der Korrektheit und der
Vollst¨ andigkeit kennen
– Vollst¨ andigkeit und Korrektheit von
Axiomensystemen beweisen k¨ onnen
8 E/R-Modelle
– Das E/R-Modell kennen
– Daten mittels des E/R-Modells modellieren
k¨ onnen
– Konzept der Relationen-Algebra kennen
– Anfragen mittels Operationen formulieren
k¨ onnen
– Joins verstehen und anwenden k¨ onnen
9 Pr¨ adikatenlogik: Syntax I
– Die ge¨ anderte Bedeutung von Variablen
verstehen
– Die Begriffe Relationssymbol, Quantor und
Formel kennen und anwenden k¨ onnen
– Den Unterschied zwischen freien und
gebundenen Variablen kennen
– Nat¨urlichsprachliche Problemstellungen
pr¨ adikatenlogisch formulieren k¨ onnen
10 Pr¨ adikatenlogik: Semantik I
– Das Konzept der relationalen logischen Struktur
verstehen
– Den Begriff des Modells verstehen
– Den Begriff der Theorie kennen
– Den Folgerungs- und ¨ Aquivalenzbegriff der
Pr¨ adikatenlogik verstehen
11 Pr¨ adikatenlogik: Syntax und
Semantik II
– Die Begriffe Konstante, Funktion und Term
kennen und anwenden k¨ onnen
– Das Konzept der logischen Signatur kennen und
anwenden k¨ onnen
– Das Konzept der Substitution kennen und
anwenden k¨ onnen
12 Pr¨ adikatenlogik: Kalk¨ule
– Hilbertkalk¨ul f¨ur die Pr¨ adikatenlogik kennen
– Aussage des G¨ odel’schen Vollst¨ andigkeitssatzes
verstehen
– Bedeutung des Satzes f¨ur die Mathematik und
die Informatik einsch¨ atzen k¨ onnen
Stand vom 9. Oktober 2006
www.tcs.uni-luebeck.de/pages/tantau/teaching.html – Lehre von Till Tantau
UN
I V E R S I T ¨ A T ZU L ¨ UB E CK
I N
S
T
I T
U
T
F ¨ U R
T H E OR E T I S C H E
I N
F O
R
M
A
T
I K
Theoretical
Computer
Science
1
216
26 Pattern Library
\usepgflibrary{patterns} % L
A
T
E
X and plain T
E
X and pure pgf
\usepgflibrary[patterns] % ConT
E
Xt and pure pgf
\usetikzlibrary{patterns} % L
A
T
E
X and plain T
E
X when using Tik Z
\usetikzlibrary[patterns] % ConT
E
Xt when using Tik Z
The package defines patterns for filling areas.
Pattern name Example
horizontal lines
vertical lines
north east lines
north west lines
grid
crosshatch
dots
crosshatch dots
fivepointed stars
sixpointed stars
bricks
217
27 Petri-Net Drawing Library
\usetikzlibrary{petri} % L
A
T
E
X and plain T
E
X
\usetikzlibrary[petri] % ConT
E
Xt
This packages provides shapes and styles for drawing Petri nets.
27.1 Places
The package defines a style for drawing places of Petri nets.
• style=place This style indicates that a node is a place of a Petri net. Usually, the text of the node
should be empty since places do not contain any text. You should use the label option to add
text outside the node like its name or its capacity. You should use the tokens options, explained in
Section 27.3, to add tokens inside the place.
p
1
p
2
≥ 1
\begin{tikzpicture}
\node[place,label=above:$p_1$,tokens=2] (p1) {};
\node[place,label=below:$p_2\ge1$,right of=p1] (p2) {};
\end{tikzpicture}
• style=every place This stype is envoked by the style place. To change the appearance of places,
you can change this style.
p
1
3
2 9
p
2
≥ 1
\tikzstyle{every place}=[draw=blue,fill=blue!20,thick,minimum size=9mm]
\begin{tikzpicture}
\node[place,tokens=7,label=above:$p_1$] (p1) {};
\node[place,structured tokens={3,2,9},
label=below:$p_2\ge1$,right of=p1] (p2) {};
\end{tikzpicture}
27.2 Transitions
Transitions are also nodes. They should be drawn using the following style:
• style=transition This style indicates that a node is a transition. As for places, the text of a transition
should be empty and the label option should be used for adding labels.
To connect a transition to places, you can use the edge command as in the following example:
p
1
p
2
≥ 1
t
1
2
\begin{tikzpicture}[node distance=2cm]
\node[place,tokens=2,label=above:$p_1$] (p1) {};
\node[place,label=above:$p_2\ge1$,right of=p1] (p2) {};
\node[transition,below right of=p1,label=below:$t_1$] {}
edge[pre] (p1)
edge[post] node[auto] {2} (p2);
\end{tikzpicture}
• style=every transition This style is envoked by the style transition.
• style=pre This style can be used with paths leading from a transition to a place to indicate that the
place is in the pre-set of the transition. By default, this style is <-,shorten <=1pt, but feel free to
redefine it.
• style=post This style is also used with paths leading from a transition to a place, but this time the
place is in the post-set of the transition. Again, feel free to redefine it.
• style=pre and post This style is to be used to indicate that a place is both in the pre- and post-set
of a transition.
218
27.3 Tokens
Interestingly, the most complicated aspect of drawing Petri nets in TikZ turns out to be the placement of
tokens.
Let us start with a single token. They are also nodes and there is a simple style for typesetting them.
• style=token This style indicates that a node is a token. By default, this causes the node to be a
small black circle. Unlike places and transitions, it does make sense to provide text for the token node.
Such text will be typeset in a tiny font and in white on black (naturally, you can easily change this by
setting the style every token).
p
1
p
2
y
\begin{tikzpicture}
\node[place,label=above:$p_1$] (p1) {};
\node[token] at (p1) {};
\node[place,label=above:$p_2$,right of=p1] (p2) {};
\node[token] at (p2) {$y$};
\end{tikzpicture}
In the above example, it is bothersome that we need an extra command for the token node. Worse, when
we have two tokens on a node, it is difficult to place both nodes inside the node without overlap.
The Petri net library offers a solution to this problem: The children are tokens style.
• style=children are tokens The idea behind this style is to use trees mechanism for placing tokens.
Every token lying on a place is treated as a child of the node. Normally this would have the effect
that the tokens are placed below the place and they would be connected to the place by an edge. The
children are tokens style, however, redefines the growth function of trees such that it places the
children next to each other inside (or, rather, on top) of the place node. Additionally, the edge from
the parent node is not drawn.
p
1
1
2 3
\begin{tikzpicture}
\node[place,label=above:$p_1$] {}
[children are tokens]
child {node [token] {1}}
child {node [token] {2}}
child {node [token] {3}};
\end{tikzpicture}
In detail, what happens is the following: Tree growth functions tell TikZ where it should place the
children of nodes. These functions get passed the number of children that a node has an the number
of the child that should be placed. The special tree growth function for tokens has a special mapping
for each possible number of children up to nine children. This mapping decides for each child where
it should be placed on top of the place. For example, a single child is placed directly on top of the
place. Two children are placed next to each other, separated by the token distance. Three children
are placed in a triangle whose side lengths are token distance; and so on up to nine tokens. If you
wish to place more than nice tokens on a place, you will have to write your own placement code.
p
2
1 2
2 1
\begin{tikzpicture}
\node[place,label=above:$p_2$] {}
[children are tokens]
child {node [token] {1}}
child {node [token,fill=red] {2}}
child {node [token,fill=red] {2}}
child {node [token] {1}};
\end{tikzpicture}
• token distance=¸distance) This specifies the distance between the centers of the tokens in the ar-
rangements of the option children are tokens.
p
3
\begin{tikzpicture}
\node[place,label=above:$p_3$] {}
[children are tokens,token distance=1.1ex]
child {node [token] {}}
child {node [token,red] {}}
child {node [token,blue] {}}
child {node [token] {}};
\end{tikzpicture}
219
The children are tokens options gives you a lot of flexibility, but it is a bit cumbersome to use. For
this reason there are some options that help in standard situations. They all use children are tokens
internally, so any change to, say, the every tokens style will affect how these options depict tokens.
• tokens=¸number) This option is given to a place node, not to a token node. The effect of this option
is to add ¸number) many child nodes to the place, each having the style token. Thus, the following
two pieces of codes have the same effect:
\tikz
\node[place] {}
[children are tokens]
child {node [token] {}}
child {node [token] {}}
child {node [token] {}};
\tikz
\node[place,tokens=3] {};
It is legal to say tokens=0, no tokens are drawn in this case. This option does not handle ten or more
tokens correctly. If you need this many tokens, you will have to program your own code.
\begin{tikzpicture}
\tikzstyle{every place}=[minimum size=9mm]
\foreach \x/\y/\tokennumber in {0/2/1,1/2/2,2/2/3,
0/1/4,1/1/5,2/1/6,
0/0/7,1/0/8,2/0/9}
\node [place,tokens=\tokennumber] at (\x,\y) {};
\end{tikzpicture}
• colored tokens=¸color list) This option, which must also be given when a place node is being created,
gets a list of colors as parameter. It will then add as many tokens to the place are in this list, each
colored with the corresponding color in the list.
\tikz \node[place,colored tokens={black,black,red,blue}] {};
• structured tokens=¸token texts) This option, which must again be passed to a place, gets a list texts
for tokens. For each text, another token will be added to the place.
x
y z
\tikz \node[place,structured tokens={$x$,$y$,$z$}] {};
1 1 2
1
2 3
1 2
3 4
1
2 3
4 5
1 2 3
4 5 6
1
2
3
4
5
6
7
1 2
3 4 5
6 7 8
1 2 3
4 5 6
7 8 9
\begin{tikzpicture}
\tikzstyle{every place}=[minimum size=9mm]
\foreach \x/\y/\tokennumber in {0/2/1,1/2/2,2/2/3,
0/1/4,1/1/5,2/1/6,
0/0/7,1/0/8,2/0/9}
\node [place,structured tokens={1,...,\tokennumber}] at (\x,\y) {};
\end{tikzpicture}
If you use lots of structured tokens, consider redefining the every token style so that the tokens are
larger.
220
28 Plot Handler Library
\usepgflibrary{plothandlers} % L
A
T
E
X and plain T
E
X and pure pgf
\usepgflibrary[plothandlers] % ConT
E
Xt and pure pgf
\usetikzlibrary{plothandlers} % L
A
T
E
X and plain T
E
X when using Tik Z
\usetikzlibrary[plothandlers] % ConT
E
Xt when using Tik Z
This library packages defines additional plot handlers, see Section 56.3 for an introduction to plot
handlers. The additional handlers are described in the following.
This library is loaded automatically by TikZ.
28.1 Curve Plot Handlers
\pgfplothandlercurveto
This handler will issue a \pgfpathcurveto command for each point of the plot, except possibly
for the first. As for the line-to handler, what happens with the first point can be specified using
\pgfsetmovetofirstplotpoint or \pgfsetlinetofirstplotpoint.
Obviously, the \pgfpathcurveto command needs, in addition to the points on the path, some control
points. These are generated automatically using a somewhat “dumb” algorithm: Suppose you have
three points x, y, and z on the curve such that y is between x and z:
x
y
z
\begin{tikzpicture}
\draw[gray] (0,0) node {x} (1,1) node {y} (2,.5) node {z};
\pgfplothandlercurveto
\pgfplotstreamstart
\pgfplotstreampoint{\pgfpoint{0cm}{0cm}}
\pgfplotstreampoint{\pgfpoint{1cm}{1cm}}
\pgfplotstreampoint{\pgfpoint{2cm}{.5cm}}
\pgfplotstreamend
\pgfusepath{stroke}
\end{tikzpicture}
In order to determine the control points of the curve at the point y, the handler computes the vector
z −x and scales it by the tension factor (see below). Let us call the resulting vector s. Then y +s and
y − s will be the control points around y. The first control point at the beginning of the curve will be
the beginning itself, once more; likewise the last control point is the end itself.
\pgfsetplottension{¸value)}
Sets the factor used by the curve plot handlers to determine the distance of the control points from
the points they control. The higher the curvature of the curve points, the higher this value should be.
A value of 1 will cause four points at quarter positions of a circle to be connected using a circle. The
default is 0.5.
x
y
z
\begin{tikzpicture}
\draw[gray] (0,0) node {x} (1,1) node {y} (2,.5) node {z};
\pgfsetplottension{0.75}
\pgfplothandlercurveto
\pgfplotstreamstart
\pgfplotstreampoint{\pgfpoint{0cm}{0cm}}
\pgfplotstreampoint{\pgfpoint{1cm}{1cm}}
\pgfplotstreampoint{\pgfpoint{2cm}{0.5cm}}
\pgfplotstreamend
\pgfusepath{stroke}
\end{tikzpicture}
\pgfplothandlerclosedcurve
This handler works like the curve-to plot handler, only it will add a new part to the current path that
is a closed curve through the plot points.
221
x
y
z
\begin{tikzpicture}
\draw[gray] (0,0) node {x} (1,1) node {y} (2,.5) node {z};
\pgfplothandlerclosedcurve
\pgfplotstreamstart
\pgfplotstreampoint{\pgfpoint{0cm}{0cm}}
\pgfplotstreampoint{\pgfpoint{1cm}{1cm}}
\pgfplotstreampoint{\pgfpoint{2cm}{0.5cm}}
\pgfplotstreamend
\pgfusepath{stroke}
\end{tikzpicture}
28.2 Comb Plot Handlers
There are three “comb” plot handlers. There name stems from the fact that the plots they produce look
like “combs” (more or less).
\pgfplothandlerxcomb
This handler converts each point in the plot stream into a line from the y-axis to the point’s coordinate,
resulting in a “horizontal comb.”
x
y
z
\begin{tikzpicture}
\draw[gray] (0,0) node {x} (1,1) node {y} (2,.5) node {z};
\pgfplothandlerxcomb
\pgfplotstreamstart
\pgfplotstreampoint{\pgfpoint{0cm}{0cm}}
\pgfplotstreampoint{\pgfpoint{1cm}{1cm}}
\pgfplotstreampoint{\pgfpoint{2cm}{0.5cm}}
\pgfplotstreamend
\pgfusepath{stroke}
\end{tikzpicture}
\pgfplothandlerycomb
This handler converts each point in the plot stream into a line from the x-axis to the point’s coordinate,
resulting in a “vertical comb.”
This handler is useful for creating “bar diagrams.”
x
y
z
\begin{tikzpicture}
\draw[gray] (0,0) node {x} (1,1) node {y} (2,.5) node {z};
\pgfplothandlerycomb
\pgfplotstreamstart
\pgfplotstreampoint{\pgfpoint{0cm}{0cm}}
\pgfplotstreampoint{\pgfpoint{1cm}{1cm}}
\pgfplotstreampoint{\pgfpoint{2cm}{0.5cm}}
\pgfplotstreamend
\pgfusepath{stroke}
\end{tikzpicture}
\pgfplothandlerpolarcomb
This handler converts each point in the plot stream into a line from the origin to the point’s coordinate.
x
y
z
\begin{tikzpicture}
\draw[gray] (0,0) node {x} (1,1) node {y} (2,.5) node {z};
\pgfplothandlerpolarcomb
\pgfplotstreamstart
\pgfplotstreampoint{\pgfpoint{0cm}{0cm}}
\pgfplotstreampoint{\pgfpoint{1cm}{1cm}}
\pgfplotstreampoint{\pgfpoint{2cm}{0.5cm}}
\pgfplotstreamend
\pgfusepath{stroke}
\end{tikzpicture}
28.3 Mark Plot Handler
\pgfplothandlermark{¸mark code)}
222
This command will execute the ¸mark code) for some points of the plot, but each time the coordinate
transformation matrix will be setup such that the origin is at the position of the point to be plotted.
This way, if the ¸mark code) draws a little circle around the origin, little circles will be drawn at some
point of the plot.
By default, a mark is drawn at all points of the plot. However, two parameters r and p influence this.
First, only every rth mark is drawn. Second, the first mark drawn is the pth. These parameters can be
influenced using the commands below.
x
y
z
\begin{tikzpicture}
\draw[gray] (0,0) node {x} (1,1) node {y} (2,.5) node {z};
\pgfplothandlermark{\pgfpathcircle{\pgfpointorigin}{4pt}\pgfusepath{stroke}}
\pgfplotstreamstart
\pgfplotstreampoint{\pgfpoint{0cm}{0cm}}
\pgfplotstreampoint{\pgfpoint{1cm}{1cm}}
\pgfplotstreampoint{\pgfpoint{2cm}{0.5cm}}
\pgfplotstreamend
\pgfusepath{stroke}
\end{tikzpicture}
Typically, the ¸code) will be \pgfuseplotmark{¸plot mark name)}, where ¸plot mark name) is the name
of a predefined plot mark.
\pgfsetplotmarkrepeat{¸repeat)}
Sets the r parameter to ¸repeat), that is, only every rth mark will be drawn.
\pgfsetplotmarkphase{¸phase)}
Sets the p parameter to ¸phase), that is, the first mark to be drawn is the pth, followed by the (p+r)th,
then the (p + 2r)th, and so on.
\pgfplothandlermarklisted{¸mark code)}{¸index list)}
This command works similar to the previous one. However, marks will only be placed at those indices
in the given ¸index list). The syntax for the list is the same as for the \foreach statement. For
example, if you provide the list 1,3,...,25, a mark will be placed only at every second point. Similarly,
1,2,4,8,16,32 yields marks only at those points that are powers of two.
x
y
z
\begin{tikzpicture}
\draw[gray] (0,0) node {x} (1,1) node {y} (2,.5) node {z};
\pgfplothandlermarklisted
{\pgfpathcircle{\pgfpointorigin}{4pt}\pgfusepath{stroke}}
{1,3}
\pgfplotstreamstart
\pgfplotstreampoint{\pgfpoint{0cm}{0cm}}
\pgfplotstreampoint{\pgfpoint{1cm}{1cm}}
\pgfplotstreampoint{\pgfpoint{2cm}{0.5cm}}
\pgfplotstreamend
\pgfusepath{stroke}
\end{tikzpicture}
\pgfuseplotmark{¸plot mark name)}
Draws the given ¸plot mark name) at the origin. The ¸plot mark name) must previously have been
declared using \pgfdeclareplotmark.
x
y
z
\begin{tikzpicture}
\draw[gray] (0,0) node {x} (1,1) node {y} (2,.5) node {z};
\pgfplothandlermark{\pgfuseplotmark{pentagon}}
\pgfplotstreamstart
\pgfplotstreampoint{\pgfpoint{0cm}{0cm}}
\pgfplotstreampoint{\pgfpoint{1cm}{1cm}}
\pgfplotstreampoint{\pgfpoint{2cm}{0.5cm}}
\pgfplotstreamend
\pgfusepath{stroke}
\end{tikzpicture}
223
\pgfdeclareplotmark{¸plot mark name)}{¸code)}
Declares a plot mark for later used with the \pgfuseplotmark command.
x
y
z
\pgfdeclareplotmark{my plot mark}
{\pgfpathcircle{\pgfpoint{0cm}{1ex}}{1ex}\pgfusepathqstroke}
\begin{tikzpicture}
\draw[gray] (0,0) node {x} (1,1) node {y} (2,.5) node {z};
\pgfplothandlermark{\pgfuseplotmark{my plot mark}}
\pgfplotstreamstart
\pgfplotstreampoint{\pgfpoint{0cm}{0cm}}
\pgfplotstreampoint{\pgfpoint{1cm}{1cm}}
\pgfplotstreampoint{\pgfpoint{2cm}{0.5cm}}
\pgfplotstreamend
\pgfusepath{stroke}
\end{tikzpicture}
\pgfsetplotmarksize{¸dimension)}
This command sets the T
E
X dimension \pgfplotmarksize to ¸dimension). This dimension is a “rec-
ommendation” for plot mark code at which size the plot mark should be drawn; plot mark code may
choose to ignore this ¸dimension) altogether. For circles, ¸dimension) should be the radius, for other
shapes it should be about half the width/height.
The predefined plot marks all take this dimension into account.
x
y
z
\begin{tikzpicture}
\draw[gray] (0,0) node {x} (1,1) node {y} (2,.5) node {z};
\pgfsetplotmarksize{1ex}
\pgfplothandlermark{\pgfuseplotmark{*}}
\pgfplotstreamstart
\pgfplotstreampoint{\pgfpoint{0cm}{0cm}}
\pgfplotstreampoint{\pgfpoint{1cm}{1cm}}
\pgfplotstreampoint{\pgfpoint{2cm}{0.5cm}}
\pgfplotstreamend
\pgfusepath{stroke}
\end{tikzpicture}
\pgfplotmarksize
A T
E
X dimension that is a “recommendation” for the size of plot marks.
The following plot marks are predefined (the filling color has been set to yellow):
\pgfuseplotmark{*}
\pgfuseplotmark{x}
\pgfuseplotmark{+}
224
29 Plot Mark Library
\usepgflibrary{plotmarks} % L
A
T
E
X and plain T
E
X and pure pgf
\usepgflibrary[plotmarks] % ConT
E
Xt and pure pgf
\usetikzlibrary{plotmarks} % L
A
T
E
X and plain T
E
X when using Tik Z
\usetikzlibrary[plotmarks] % ConT
E
Xt when using Tik Z
This library defines a number of plot marks.
This library defines the following plot marks in addition to *, x, and + (the filling color has been set to
a dark yellow):
\pgfuseplotmark{-}
\pgfuseplotmark{|}
\pgfuseplotmark{o}
\pgfuseplotmark{asterisk}
\pgfuseplotmark{star}
\pgfuseplotmark{oplus}
\pgfuseplotmark{oplus*}
\pgfuseplotmark{otimes}
\pgfuseplotmark{otimes*}
\pgfuseplotmark{square}
\pgfuseplotmark{square*}
\pgfuseplotmark{triangle}
\pgfuseplotmark{triangle*}
\pgfuseplotmark{diamond}
\pgfuseplotmark{diamond*}
\pgfuseplotmark{pentagon}
\pgfuseplotmark{pentagon*}
225
30 Shape Library
In addition to the standard shapes rectangle, circle and coordinate, there exist a number of additional
shapes defined in different shape libraries. In the present section, these shapes are described.
\usepgflibrary{shapes} % L
A
T
E
X and plain T
E
X and pure pgf
\usepgflibrary[shapes] % ConT
E
Xt and pure pgf
\usetikzlibrary{shapes} % L
A
T
E
X and plain T
E
X when using Tik Z
\usetikzlibrary[shapes] % ConT
E
Xt when using Tik Z
This library packages just includes all of the libraries defined in the following. Note that it includes only
those libraries starting with shapes., more special-purpose libraries are described in dedicated sections.
30.1 Geometric Shapes
\usepgflibrary{shapes.geometric} % L
A
T
E
X and plain T
E
X and pure pgf
\usepgflibrary[shapes.geometric] % ConT
E
Xt and pure pgf
\usetikzlibrary{shapes.geometric} % L
A
T
E
X and plain T
E
X when using Tik Z
\usetikzlibrary[shapes.geometric] % ConT
E
Xt when using Tik Z
This library defines different shapes that correspond to basic geometric objects like ellipses or polygons.
Shape diamond
This shape is a diamond tightly fitting the text box. The ratio between width and height is 1 by default,
but can be changed by setting the shape aspect ratio (using the aspect option of TikZ). The following
figure shows the anchors this shape defines; the anchors 10 and 130 are example of border anchors.
Diamond
(s.north west)
(s.north)
(s.north east)
(s.west)
(s.center)
(s.east)
(s.mid)
(s.base)
(s.south west)
(s.south)
(s.south east)
(s.text)
(s.10)
(s.130)
226
\Huge
\begin{tikzpicture}
\node[name=s,shape=diamond,style=shape example] {Diamond\vrule width 1pt height 2cm};
\foreach \anchor/\placement in
{north west/above left, north/above, north east/above right,
west/left, center/above, east/right,
mid/above,
base/below,
south west/below left, south/below, south east/below right,
text/left, 10/right, 130/above}
\draw[shift=(s.\anchor)] plot[mark=x] coordinates{(0,0)}
node[\placement] {\scriptsize\texttt{(s.\anchor)}};
\end{tikzpicture}
Shape ellipse
This shape is an ellipse tightly fitting the text box, if no inner separation is given. The following figure
shows the anchors this shape defines; the anchors 10 and 130 are example of border anchors.
Ellipse
(s.north west)
(s.north)
(s.north east)
(s.west)
(s.center)
(s.east)
(s.mid west)
(s.mid)
(s.mid east)
(s.base west)
(s.base)
(s.base east)
(s.south west)
(s.south)
(s.south east)
(s.text)
(s.10)
(s.130)
\Huge
\begin{tikzpicture}
\node[name=s,shape=ellipse,style=shape example] {Ellipse\vrule width 1pt height 2cm};
\foreach \anchor/\placement in
{north west/above left, north/above, north east/above right,
west/left, center/above, east/right,
mid west/right, mid/above, mid east/left,
base west/left, base/below, base east/right,
south west/below left, south/below, south east/below right,
text/left, 10/right, 130/above}
\draw[shift=(s.\anchor)] plot[mark=x] coordinates{(0,0)}
node[\placement] {\scriptsize\texttt{(s.\anchor)}};
\end{tikzpicture}
Shape regular polygon
This shape is a regular polygon, which, by default, is drawn so that a side (rather than a corner) is
always at the bottom.
5 6 7 8
\begin{tikzpicture}
\foreach \i in {5,...,8}
\node[regular polygon, regular polygon sides=\i, draw] at (\i,0) {\i};
\end{tikzpicture}
Two points should be remembered regarding the dimensions of the polygon border. Firstly, the border
is constructed using the incircle, that is, the circle that touches every side of the polygon border. The
radius of the incircle is calculated to tightly fit the node contents.
227
\begin{tikzpicture}
\foreach \a in {3,...,7}{
\draw[gray!50] (\a*2,0) circle(0.5cm);
\node[regular polygon, regular polygon sides=\a, inner sep=0cm, draw] at (\a*2,0)
{\tikz\fill[red!50] rectangle(0.707cm,0.707cm);};
}
\end{tikzpicture}
Secondly, if the node is enlarged to any specified minimum size, width or height, this is interpreted as
the diameter of the the circumcircle, that is, the circle that passes through all the corners of the polygon
border.
\begin{tikzpicture}
\foreach \a in {3,...,7}{
\draw[gray!50] (\a*2,0) circle(0.5cm);
\node[regular polygon, regular polygon sides=\a, minimum size=1cm, draw] at (\a*2,0) {};
}
\end{tikzpicture}
There are pgf commands and TikZ options to set the number of sides for the polygon, and the rotation
of the polygon border. The pgf commands are as follows:
\pgfsetpolygonsides{¸integer)}
Set the number of sides for the polygon.
\pgfsetpolygonrotate{¸angle)}
Rotate the border of the polygon independently of the node contents (but in addition to any
concurrent coordinate or canvas transformation).
The corresponding TikZ options are:
• regular polygon sides=¸integer) set the number of points for the star.
• regular polygon rotate=¸angle) rotate the border of the polygon independently of the node
contents.
The anchors for the regular polygon shape are shown below:
228
Regular Polygon
(s.corner 1)
(s.corner 2)
(s.corner 3) (s.corner 4)
(s.corner 5)
(s.side 1)
(s.side 2)
(s.side 3)
(s.side 4)
(s.side 5)
(s.mid)
(s.base)
(s.center)
(s.text)
(s.75)
(s.west) (s.east)
(s.north)
(s.south)
\Huge
\begin{tikzpicture}
\node[name=s, shape=regular polygon, regular polygon sides=5, style=shape example, inner sep=.5cm]
{Regular Polygon\vrule width 1pt height 2cm};
\foreach \anchor/\placement in
{corner 1/above, corner 2/above, corner 3/left, corner 4/right,
corner 5/above, side 1/left, side 2/left, side 3/below,
side 4/right, side 5/right, mid/right, base/below,
center/above, text/left, 75/above, west/above,
east/above, north/below, south/above}
\draw[shift=(s.\anchor)] plot[mark=x] coordinates{(0,0)}
node[\placement] {\scriptsize\texttt{(s.\anchor)}};
\end{tikzpicture}
Shape star
This shape is a star, which by default (minus any transformations) is drawn with the first point pointing
upwards. A star should be thought of as having an set of “inner points” and and “outer points”. These
points form the principle anchors for the star, as shown below:
229
Star
(s.inner point 1)
(s.inner point 2)
(s.inner point 3)
(s.inner point 4)
(s.inner point 5)
(s.outer point 1)
(s.outer point 2)
(s.outer point 3) (s.outer point 4)
(s.outer point 5)
(s.mid)
(s.base)
(s.center)
(s.text)
(s.75)
\Huge
\begin{tikzpicture}
\node[name=s, shape=star, star points=5, star point ratio=1.5, style=shape example, inner sep=1.5cm]
{Star\vrule width 1pt height 2cm};
\foreach \anchor/\placement in
{inner point 1/above, inner point 2/above, inner point 3/below,
inner point 4/right, inner point 5/above, outer point 1/above,
outer point 2/above, outer point 3/left, outer point 4/right,
outer point 5/above, mid/right, base/below,
center/above, text/left, 75/above}
\draw[shift=(s.\anchor)] plot[mark=x] coordinates{(0,0)}
node[\placement] {\scriptsize\texttt{(s.\anchor)}};
\end{tikzpicture}
The inner points of the border are based on the radius of the circle which tightly fits the node contents.
Any specified minimum size, width or height, is interpreted as the diameter of the circle that passes
through every outer point.
There are pgf commands and TikZ options to set various parameters for the star, such as the number
of points, the height of the points and the rotation of the star border.
The pgf commands are as follows:
\pgfsetstarpoints{¸integer)}
Sets the number of points for the star.
\pgfsetstarpointheight{¸distance)}
Sets the height of the star points. This is the distance between the inner point and outer point radii.
If the star is enlarged to some specified minimum size, the inner radius is increased to maintain the
point height.
\pgfsetstarpointratio{¸number)}
Sets the ratio between the outer point and inner point radii. If the star is enlarged to some specified
minimum size, the inner radius is increased to maintain the ratio.
\pgfsetstarrotate{¸angle)}
230
Rotates the border of the star, independently of the node contents, but subject to any coordinate
or canvas transformations.
The corresponding TikZ options are:
• star points=¸integer) set the number of points for the star.
• star point height=¸distance) set the height of the points for the star.
• star point ratio=¸number) set the ratio between the outer point radius and the inner point
radius.
• star rotate=¸angle) rotate the star shape border indepently of the node contents.
30.2 Symbol Shapes
\usepgflibrary{shapes.symbols} % L
A
T
E
X and plain T
E
X and pure pgf
\usepgflibrary[shapes.symbols] % ConT
E
Xt and pure pgf
\usetikzlibrary{shapes.symbols} % L
A
T
E
X and plain T
E
X when using Tik Z
\usetikzlibrary[shapes.symbols] % ConT
E
Xt when using Tik Z
This library defines shapes that can be used for drawing symbols like a forbidden sign or a cloud.
Shape forbidden sign
This shape places the node inside a circle with a diagonal from the lower left to the upper right added.
The circle is part of the background, the diagonal line part of the foreground path; thus, the diagonal
line is on top of the text.
Smoking
\begin{tikzpicture}
\node [forbidden sign,line width=1ex,draw=red,fill=white] {Smoking};
\end{tikzpicture}
The shape inherits all anchors from the circle shape, see also the following figure:
Forbidden
(s.north west)
(s.north)
(s.north east)
(s.west)
(s.center)
(s.east)
(s.mid west)
(s.mid)
(s.mid east)
(s.base west)
(s.base)
(s.base east)
(s.south west)
(s.south)
(s.south east)
(s.text)
(s.10)
(s.130)
231
\Huge
\begin{tikzpicture}
\node[name=s,shape=forbidden sign,style=shape example] {Forbidden\vrule width 1pt height 2cm};
\foreach \anchor/\placement in
{north west/above left, north/above, north east/above right,
west/left, center/above, east/right,
mid west/right, mid/above, mid east/left,
base west/left, base/below, base east/right,
south west/below left, south/below, south east/below right,
text/left, 10/right, 130/above}
\draw[shift=(s.\anchor)] plot[mark=x] coordinates{(0,0)}
node[\placement] {\scriptsize\texttt{(s.\anchor)}};
\end{tikzpicture}
30.3 Shapes with Multiple Text Parts
\usepgflibrary{shapes.multipart} % L
A
T
E
X and plain T
E
X and pure pgf
\usepgflibrary[shapes.multipart] % ConT
E
Xt and pure pgf
\usetikzlibrary{shapes.multipart} % L
A
T
E
X and plain T
E
X when using Tik Z
\usetikzlibrary[shapes.multipart] % ConT
E
Xt when using Tik Z
This library defines general-purpose shapes that are composed of multiple (text) parts.
Shape circle split
This shape is a multi-part shape consisting of a circle with a line in the middle. The upper part is the
main part (the text part), the lower part is the lower part.
q
1
00
\begin{tikzpicture}
\node [circle split,draw,double,fill=red!20]
{
$q_1$
\nodepart{lower}
$00$
};
\end{tikzpicture}
The shape inherits all anchors from the circle shape and defines the lower anchor in addition. See
also the following figure:
text
lower
(s.north west)
(s.north)
(s.north east)
(s.west)
(s.center)
(s.east)
(s.mid west)
(s.mid)
(s.mid east)
(s.base west)
(s.base)
(s.base east)
(s.south west)
(s.south)
(s.south east)
(s.text)
(s.lower)
(s.130)
232
\Huge
\begin{tikzpicture}
\node[name=s,shape=circle split,style=shape example] {text\nodepart{lower}lower};
\foreach \anchor/\placement in
{north west/above left, north/above, north east/above right,
west/left, center/below, east/right,
mid west/right, mid/above, mid east/left,
base west/left, base/below, base east/right,
south west/below left, south/below, south east/below right,
text/left, lower/left, 130/above}
\draw[shift=(s.\anchor)] plot[mark=x] coordinates{(0,0)}
node[\placement] {\scriptsize\texttt{(s.\anchor)}};
\end{tikzpicture}
30.4 Miscellaneous Shapes
\usepgflibrary{shapes.misc} % L
A
T
E
X and plain T
E
X and pure pgf
\usepgflibrary[shapes.misc] % ConT
E
Xt and pure pgf
\usetikzlibrary{shapes.misc} % L
A
T
E
X and plain T
E
X when using Tik Z
\usetikzlibrary[shapes.misc] % ConT
E
Xt when using Tik Z
This library defines general-purpose shapes that do not fit in the previous categories.
Shape cross out
This shape “crosses out” the node. Its foreground path are simply two diagonal lines that between the
corners of the node’s bounding box. Here is an example:
cross out
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\node [cross out,draw=red] at (1.5,1) {cross out};
\end{tikzpicture}
A useful application is inside text as in the following example:
Cross me out!
Cross \tikz[baseline] \node [cross out,draw,anchor=text] {me}; out!
This shape inherits all anchors from the rectangle shape, see also the following figure:
cross out
(s.north west) (s.north) (s.north east)
(s.west)
(s.center)
(s.east)
(s.mid west)
(s.mid)
(s.mid east)
(s.base west)
(s.base)
(s.base east)
(s.south west) (s.south) (s.south east)
(s.text)
(s.10)
(s.130)
\Huge
\begin{tikzpicture}
\node[name=s,shape=cross out,style=shape example] {cross out\vrule width 1pt height 2cm};
\foreach \anchor/\placement in
{north west/above left, north/above, north east/above right,
west/left, center/above, east/right,
mid west/right, mid/above, mid east/left,
base west/left, base/below, base east/right,
south west/below left, south/below, south east/below right,
text/left, 10/right, 130/above}
\draw[shift=(s.\anchor)] plot[mark=x] coordinates{(0,0)}
node[\placement] {\scriptsize\texttt{(s.\anchor)}};
\end{tikzpicture}
233
Shape strike out
This shape is idential to the cross out shape, only its foreground path consists of a single line from
the lower left to the upper right.
Strike me out!
Strike \tikz[baseline] \node [strike out,draw,anchor=text] {me}; out!
See the cross out shape for the anchors.
234
31 Snake Library
\usepgflibrary{snakes} % L
A
T
E
X and plain T
E
X and pure pgf
\usepgflibrary[snakes] % ConT
E
Xt and pure pgf
\usetikzlibrary{snakes} % L
A
T
E
X and plain T
E
X when using Tik Z
\usetikzlibrary[snakes] % ConT
E
Xt when using Tik Z
This library package defines basic snakes. Section 11.2.3 explains how snakes are used in TikZ, Section 46
explains how new snakes can be defined.
The snakes are influenced by the current values of parameters like \pgfsnakesegmentamplitude. Only
this parameter and \pgfsnakesegmentlength are proper T
E
X dimensions, all other parameters are T
E
X
macros.
In TikZ, each parameter can be set using an option having the parameters name minus the \pgfsnake
part.
Snake bent
This snake adds a slightly bent line from the start to the target. The amplitude of the bent is given by
the segement amplitude (and amplitude of zero gives a straight line).
• \pgfsnakesegmentamplitude determines the amplitude of the bent.
• \pgfsnakesegmentaspect determines how tight the bent is. A good value is around 0.3.
A
B
\begin{tikzpicture}[segment aspect=.3]
\node[circle,draw] (A) at (.5,.5) {A};
\node[circle,draw] (B) at (3,1.5) {B};
\draw[->,snake=bent,raise snake=2pt] (A) -- (B);
\draw[->,snake=bent,raise snake=2pt] (B) -- (A);
\draw [snake=bent,mirror snake] (0,0) rectangle (3.5,2);
\end{tikzpicture}
Snake border
This snake adds straight lines the path that are at a specific angle to the line toward the target. The
idea is to add these little lines to indicate the “border” or an area. The following parameters influence
the snake:
• \pgfsnakesegmentlength determines the distance between consecutive ticks.
• \pgfsnakesegmentamplitude determines the length of the ticks.
• \pgfsnakesegmentangle determines the angle between the ticks and the line toward the target.
\tikz{\draw (0,0) rectangle (3,1)
[snake=border,segment angle=-45] (0,0) rectangle (3,1);}
Snake brace
This snake adds a long brace to the path. The left and right end of the brace will be exactly on the
start and endpoint of the snake. The following parameters influence the snake:
• \pgfsnakesegmentamplitude determines how much the brace rises above the path.
• \pgfsnakesegmentaspect determines the fraction of the total length where the “middle part” of
the brace will be.
\tikz{\draw[snake=brace,segment aspect=0.25] (0,0) -- (3,0);}
Snake bumps
This snake consists of little half ellipses. The following parameters influence the snake:
• \pgfsnakesegmentamplitude determines the height of the half ellipse.
235
• \pgfsnakesegmentlength determines the width of the half ellipse.
\tikz{\draw[snake=bumps] (0,0) -- (3,0);}
Snake coil
This snake adds a coil to the path. To understand how this works, imagine a three-dimensional spring.
The spring’s axis points along the line toward the target. Then, we “view” the spring from a certain
angle. If we look “straight from the side” we will see a perfect sine curve, if we look “more from the
front” we will see a coil. The following parameters influence the snake:
• \pgfsnakesegmentamplitude determines how much the coil rises above the path and falls below
it. Thus, this is the radius of the coil.
• \pgfsnakesegmentlength determines the distance between two consecutive “curls.” Thus, when
the spring is see “from the side” this will be the wave length of the sine curve.
• \pgfsnakesegmentaspect determines the “viewing direction.” A value of 0 means “looking from
the side” and a value of 0.5, which is the default, means “look more from the front.”
\begin{tikzpicture}[segment amplitude=10pt]
\draw[snake=coil] (0,1) -- (3,1);
\draw[snake=coil,segment aspect=0] (0,0) -- (3,0);
\end{tikzpicture}
Snake expanding waves
This snake adds arcs to the path that get bigger along the line towards the target. The following
parameters influence the snake:
• \pgfsnakesegmentlength determines the distance between consecutive arcs.
• \pgfsnakesegmentangle determines the opening angle below and above the path. Thus, the total
opening angle is twice this angle.
\tikz{\draw[snake=expanding waves] (0,0) -- (3,0);}
Snake saw
This snake looks like the blade of a saw. The following parameters influence the snake:
• \pgfsnakesegmentamplitude determines how much each spike raises above the straight line.
• \pgfsnakesegmentlength determines the length each spike.
\tikz{\draw[snake=saw] (0,0) -- (3,0);}
Snake snake
This snake is the “architypical” snake: It looks like a snake seen from above. More precisely, the snake
is a sine wave with a “softened” start and ending. The following parameters influence the snake:
• \pgfsnakesegmentamplitude determines the sine wave’s amplitude.
• \pgfsnakesegmentlength determines the sine wave’s wave length.
\tikz{\draw[snake=snake] (0,0) -- (3,0);}
236
Snake ticks
This snake adds straight lines the path that are orthogonal to the line toward the target. The following
parameters influence the snake:
• \pgfsnakesegmentlength determines the distance between consecutive ticks.
• \pgfsnakesegmentamplitude determines half the length of the ticks.
\tikz{\draw[snake=ticks] (0,0) -- (3,0);}
Snake triangles
This snake adds triangles to the path that point toward the target. The following parameters influence
the snake:
• \pgfsnakesegmentlength determines the distance between consecutive triangles.
• \pgfsnakesegmentamplitude determines half the length of the triangle side that is orthogonal to
the path.
• \pgfsnakesegmentobjectlength determines the height of the triangle.
\tikz{\draw[snake=triangles] (0,0) -- (3,0);}
Snake crosses
This snake adds (diagonal) crosses to the path. The following parameters influence the snake:
• \pgfsnakesegmentlength determines the distance between consecutive crosses.
• \pgfsnakesegmentamplitude determines half the hieght of each cross.
• \pgfsnakesegmentobjectlength determines width of each cross.
\tikz{\draw[snake=crosses] (0,0) -- (3,0);}
Snake waves
This snake adds arcs to the path that have a constant size. The following parameters influence the
snake:
• \pgfsnakesegmentlength determines the distance between consecutive arcs.
• \pgfsnakesegmentangle determines the opening angle below and above the path. Thus, the total
opening angle is twice this angle.
• \pgfsnakesegmentamplitude determines the radius of each arc.
\tikz{\draw[snake=waves] (0,0) -- (3,0);}
Snake zigzag
This snake looks like a zig-zag line. The following parameters influence the snake:
• \pgfsnakesegmentamplitude determines how much the zig-zag lines raises above and falls below
a straight line to the target point.
• \pgfsnakesegmentlength determines the length of a complete “up-down” cycle.
\tikz{\draw[snake=zigzag] (0,0) -- (3,0);}
237
32 To Path Library
\usetikzlibrary{topaths} % L
A
T
E
X and plain T
E
X
\usetikzlibrary[topaths] % ConT
E
Xt
This library provides predefined to paths for use with the to path operation. After loading this package,
you can say for instance to [loop] to add a loop to a node.
This library is loaded automatically by TikZ, so you do not need to load it yourself.
32.1 Straight Lines
The following style installs a to path that is simply a straight line from the start coordinate to the target
coordinate.
• style=line to causes a straight line to be added to the path upon a to or an edge operation.
\tikz {\draw (0,0) to[line to] (1,0);}
32.2 Curves
The curve to style causes the to path to be set to a curve. The exact way this curve looks can be influenced
via a number of options.
• style=curve to Specifies that the to path should be a curve. This curve will leave the start coordi-
nate at a certain angle, which can be specified using the out option. It reaches the target coordinate
also at a certain angle, which is specified using the in option. The control points of the curve are at a
certain distance that is computed in different ways, depending on which options are set.
All of the following options implictly cause the curve to style to be installed.
• out=¸angle) The angle at which the curve leaves the start coordinate. If the start coordinate is a node,
the start coordinate is the point on the border of the node at the given ¸angle). The control point will,
thus, lie at a certain distance in the direction ¸angle) from the start coordinate.
\begin{tikzpicture}[out=45,in=135]
\draw (0,0) to (1,0)
(0,0) to (2,0)
(0,0) to (3,0);
\end{tikzpicture}
• in=¸angle) The angle at which the curve reaches the target coordinate.
• relative=¸true or false) This option tells TikZ whether the in and out angles should be considered
absolute or relative. Absolute means that an out angle of 30

means that the curve leaves the start
coordinate at an angle of 30

relative to the paper (unless, of course, further transformations have
been installed). A relative angle is, by comparison, measured relative to a straight line from the start
coordinate to the target coordinate. Thus, a relative angle of 30

means that the curve will bend to
the left from the line going straight from the start to the target. For the target, the relative coordinate
is measured in the same manner, namely relative to the line going from the start to the target. Thus,
an angle of 150

means that the curve will reach target coming slightly from the left.
\begin{tikzpicture}[out=45,in=135,relative]
\draw (0,0) to (1,0)
to (2,1)
to (2,2);
\end{tikzpicture}
238
a
b
c
\begin{tikzpicture}[out=90,in=90,relative]
\node [circle,draw] (a) at (0,0) {a};
\node [circle,draw] (b) at (1,1) {b};
\node [circle,draw] (c) at (2,2) {c};
\path (a) edge (b)
edge (c);
\end{tikzpicture}
• bend left=¸angle) This option sets out=¸angle),in=180 − ¸angle),relative. If no ¸angle) is given,
the last given bend left or bend right angle is used.
q
0 start
q
1
q
2
0
1
1
0
1
\begin{tikzpicture}[shorten >=1pt,node distance=2cm]
\node[state,initial] (q_0) {$q_0$};
\node[state] (q_1) [right of=q_0] {$q_1$};
\node[state,accepting](q_2) [right of=q_1] {$q_2$};
\path[->] (q_0) edge node [above] {0} (q_1)
edge [loop above] node {1} ()
edge [bend left] node [above] {1} (q_2)
edge [bend right] node [below] {0} (q_2)
(q_1) edge node [above] {1} (q_2);
\end{tikzpicture}
0
45
90
135
180
225
270
315
\begin{tikzpicture}
\foreach \angle in {0,45,...,315}
\node[rectangle,draw=black!50] (\angle) at (\angle:2) {\angle};
\foreach \from/\to in {0/45,45/90,90/135,135/180,
180/225,225/270,270/315,315/0}
\path (\from) edge [->,bend right=22,looseness=0.8] (\to)
edge [<-,bend left=22,looseness=0.8] (\to);
\end{tikzpicture}
• bend right=¸angle) Works like the bend left option, only the bend is to the other side.
• bend angle=¸angle) Sets the angle to be used by the bend left or bend right, but without actually
selecting the curve to or the relative option. This is useful for globally specifying a bend angle for
a whole picture.
• looseness=¸number) This number specifies how “loose” the curve will be. In detail, the following
happens: TikZ computes the distance between the start and the target coordinate (if the start and/or
target coordinate are nodes, the distance is computed between the points on their border). This
239
distance is then multiplied by a fixed factor and also by the factor ¸number). The resulting distance,
let us call it d, is then used as the distance of the control points from the start and target coordinates.
The fixed factor has been chosen in such a way that if ¸number) is 1, which is the default, if the in
and out angles differ by 90◦, then a quarter circle results:
\tikz \draw (0,0) to [out=0,in=-90] (1,1);
\tikz \draw (0,0) to [out=0,in=-90,looseness=0.5] (1,1);
• out looseness=¸number) specifies the looseness factor for the out distance only.
• in looseness=¸number) specifies the looseness factor for the in distance only.
• min distance=¸distance) If the computed distance for the start and target coordinates are below
¸distance), then ¸distance) is used instead.
• max distance=¸distance) If the computed distance for the start and target coordinates are above
¸distance), then ¸distance) is used instead.
• out min distance=¸distance) The min distance set only for the start coordinate.
• out max distance=¸distance) The max distance set only for the start coordinate.
• in min distance=¸distance) The min distance set only for the target coordinate.
• in max distance=¸distance) The max distance set only for the target coordinate.
• distance=¸distance) Set the min and max distance to the same value ¸distance). Note that this causes
any computed distance d to be ignored and ¸distance) to be used instead.
\begin{tikzpicture}[out=45,in=135,distance=1cm]
\draw (0,0) to (1,0)
(0,0) to (2,0)
(0,0) to (3,0);
\end{tikzpicture}
• out distance=¸distance) sets the min and max out distance.
• in distance=¸distance) sets the min and max in distance.
• out control=¸coordinate) This option causes the ¸coordinate) to be used as the start control point.
All computations of d are ignored. You can use a coordinate like +(1,0) to specify a point relative to
the start coordinate.
• in control=¸coordinate) This option causes the ¸coordinate) to be used as the target control point.
• controls=¸coordinate) and ¸coordinate) This option causes the ¸coordinate)s to be used as control
points.
\tikz \draw (0,0) to [controls=+(90:1) and +(90:1)] (3,0);
32.3 Loops
• style=loop This style is similar to the curve to style, but differs in the following ways: First, the
actual target coordinate is ignored and the start coordiante is used as the target coordinate. Thus, it
is allowed not to provide any target coordinate, which can be useful with unnamed nodes. Second, the
looseness is set to 8 and the min distance to 5mm. These settings result in rather nice loops when
the opening angle (difference between in and out) is 30

.
a
\begin{tikzpicture}
\node [circle,draw] {a} edge [in=30,out=60,loop] ();
\end{tikzpicture}
240
• style=loop above Sets the loop style and sets in and out angles such that loop is above the node.
Furthermore, the above option is set, which causes a node label to be placed at the correct position.
a
x
\begin{tikzpicture}
\node [circle,draw] {a} edge [loop above] node {x} ();
\end{tikzpicture}
• style=loop below works like the previous option.
• style=loop left works like the previous option.
• style=loop right works like the previous option.
• style=every loop This style is installed at the beginning of every loop. By default, it is set to
->,shorten >=1pt, but feel free to change this.
\begin{tikzpicture}
\tikzstyle{every loop}=[]
\draw (0,0) to [loop above] () to [loop right] ()
to [loop below] () to [loop left] ();
\end{tikzpicture}
241
33 Tree Library
\usetikzlibrary{trees} % L
A
T
E
X and plain T
E
X
\usetikzlibrary[trees] % ConT
E
Xt
This packages defines styles to be used when drawing trees.
33.1 Growth Functions
The package pgflibrarytikztrees defines two new growth functions. They are installed using the following
options:
• grow via three points=one child at (¸x)) and two children at (¸y)) and (¸z )) This option
installs a growth function that works as follows: If a parent node has just one child, this child is placed
at ¸x). If the parent node has two children, these are placed at ¸y) and ¸z ). If the parent node has
more than two children, the children are placed at points that are linearly extrapolated from the three
points ¸x), ¸y), and ¸z ). In detail, the position is x+
n−1
2
(y −x) +(c−1)(z −y), where n is the number
of children and c is the number of the current child (starting with 1).
The net effect of all this is that if you have a certain “linear arrangement” in mind and use this option
to specify the placement of a single child and of two children, then any number of children will be
placed correctly.
Here are some arrangements based on this growth function. We start with a simple “above” arrange-
ment:
one
two
three
four
\begin{tikzpicture}[grow via three points={%
one child at (0,1) and two children at (-.5,1) and (.5,1)}]
\node at (0,0) {one} child;
\node at (0,-1.5) {two} child child;
\node at (0,-3) {three} child child child;
\node at (0,-4.5) {four} child child child child;
\end{tikzpicture}
The next arrangement places children above, but “grows only to the right.”
one
two
three
four
\begin{tikzpicture}[grow via three points={%
one child at (0,1) and two children at (0,1) and (1,1)}]
\node at (0,0) {one} child;
\node at (0,-1.5) {two} child child;
\node at (0,-3) {three} child child child;
\node at (0,-4.5) {four} child child child child;
\end{tikzpicture}
In the final arrangement, the children are placed along a line going down and right.
242
one
two
three
four
\begin{tikzpicture}[grow via three points={%
one child at (-1,-.5) and two children at (-1,-.5) and (0,-.75)}]
\node at (0,0) {one} child;
\node at (0,-1.5) {two} child child;
\node at (0,-3) {three} child child child;
\node at (0,-4.5) {four} child child child child;
\end{tikzpicture}
These examples should make it clear how you can create new styles to arrange your children along a
line.
• style=grow cyclic This style causes the children to be arranged “on a circle.” For this, the children
are placed at distance \tikzleveldistance from the parent node, but not on a straight line, but
points on a circle. Instead of a sibling distance, there is a sibling angle that denotes the angle
between two given children.
– sibling angle=¸angle) Sets the angle between siblings in the grow cyclic style.
Note that this function will rotate the coordinate system of the children to ensure that the grandchildren
will grow in the right direction.
\begin{tikzpicture}[grow cyclic]
\tikzstyle{level 1}=[level distance=8mm,sibling angle=60]
\tikzstyle{level 2}=[level distance=4mm,sibling angle=45]
\tikzstyle{level 3}=[level distance=2mm,sibling angle=30]
\coordinate [rotate=-90] % going down
child foreach \x in {1,2,3}
{child foreach \x in {1,2,3}
{child foreach \x in {1,2,3}}};
\end{tikzpicture}
• clockwise from=¸angle) This option also cuases children to be arranged on a circle. However, the rule
for placing children is simpler thatn with the grow cyclic style: The first child is placed at ¸angle) at a
distance of \tikzleveldistance. The second child is placed at the same distance from the parent, but
at angle ¸angle) − \tikzsiblingangle. The third child is displaced by another \tikzsiblingangle
in a clockwise fashion, and so on.
Note that this function will not rotate the coordinate system.
root
30
0
−30
−60
\begin{tikzpicture}
\node {root}
[clockwise from=30,sibling angle=30]
child {node {$30$}}
child {node {$0$}}
child {node {$-30$}}
child {node {$-60$}};
\end{tikzpicture}
• counterclockwise from=¸angle) Works the same way as clockwise from, but sibling angles are
added instead of subtracted.
33.2 Edges From Parent
The following styles can be used to modify how the edges from parents are drawn:
243
• style=edge from parent fork down This style will draw a line from the parent downwards (for half
the level distance) and then on to the child using only horizontal and vertical lines.
root
left right
child child
\begin{tikzpicture}
\node {root}
[edge from parent fork down]
child {node {left}}
child {node {right}
child[child anchor=north east] {node {child}}
child {node {child}}
};
\end{tikzpicture}
• style=edge from parent fork right This style behaves similarly, only it will first draw its edge to
the right.
root
left
right
child
child
\begin{tikzpicture}
\node {root}
[edge from parent fork right,grow=right]
child {node {left}}
child {node {right}
child {node {child}}
child {node {child}}
};
\end{tikzpicture}
• style=edge from parent fork left behaves similary.
• style=edge from parent fork up behaves similary.
244
Part V
Utilities
by Till Tantau
The utility packages are not directly involved in creating graphics, but you may find them useful nonetheless.
All of them either directly depend on pgf or they are designed to work well together with pgf even though
they can be used in a stand-alone way.

3/2
0
x
2
dx
x
f(x)
1
1
1
2
2 3
1
2
2
1
4
3
x
2
\begin{tikzpicture}[scale=2]
\shade[top color=blue,bottom color=gray!50] (0,0) parabola (1.5,2.25) |- (0,0);
\draw (1.05cm,2pt) node[above] {$\displaystyle\int_0^{3/2} \!\!x^2\mathrm{d}x$};
\draw[style=help lines] (0,0) grid (3.9,3.9)
[step=0.25cm] (1,2) grid +(1,1);
\draw[->] (-0.2,0) -- (4,0) node[right] {$x$};
\draw[->] (0,-0.2) -- (0,4) node[above] {$f(x)$};
\foreach \x/\xtext in {1/1, 1.5/1\frac{1}{2}, 2/2, 3/3}
\draw[shift={(\x,0)}] (0pt,2pt) -- (0pt,-2pt) node[below] {$\xtext$};
\foreach \y/\ytext in {1/1, 2/2, 2.25/2\frac{1}{4}, 3/3}
\draw[shift={(0,\y)}] (2pt,0pt) -- (-2pt,0pt) node[left] {$\ytext$};
\draw (-.5,.25) parabola bend (0,0) (2,4) node[below right] {$x^2$};
\end{tikzpicture}
245
34 Repeating Things: The Foreach Statement
This section describes the package pgffor, which is loaded automatically by TikZ, but not by pgf:
\usepackage{pgffor} % L
A
T
E
X
\input pgffor.tex % plain T
E
X
\usemodule[pgffor] % ConT
E
Xt
This package can be used independently of pgf, but works particularly well together with pgf and
TikZ. It defines two new commands: \foreach and \breakforeach.
\foreach ¸variables) in {¸list)} ¸commands)
The syntax of this command is a bit complicated, so let us go through it step-by-step.
In the easiest case, ¸variables) is a single T
E
X-command like \x or \point. (If you want to have some
fun, you can also use active characters. If you do not know what active characters are, you are blessed.)
Still in the easiest case, ¸list) is a comma-separated list of values. Anything can be used as a value, but
numbers are most likely.
Finally, in the easiest case, ¸commands) is some T
E
X-text in curly braces.
With all these assumptions, the \foreach statement will execute the ¸commands) repeatedly, once for
every element of the ¸list). Each time the ¸commands) are executed, the ¸variable) will be set to the
current value of the list item.
[1][2][3][0]
\foreach \x in {1,2,3,0} {[\x]}
Syntax for the commands. Let us move on to a more complicated setting. The first complication
occurs when the ¸commands) are not some text in curly braces. If the \foreach statement does not
encounter an opening brace, it will instead scan everything up to the next semicolon and use this as
¸commands). This is most useful in situations like the following:
\tikz
\foreach \x in {0,1,2,3}
\draw (\x,0) circle (0.2cm);
However, the “reading till the next semicolon” is not the whole truth. There is another rule: If a
\foreach statement is directly followed by another \foreach statement, this second foreach statement
is collected as ¸commands). This allows you to write the following:
\begin{tikzpicture}
\foreach \x in {0,1,2,3}
\foreach \y in {0,1,2,3}
{
\draw (\x,\y) circle (0.2cm);
\fill (\x,\y) circle (0.1cm);
}
\end{tikzpicture}
The dots notation. The second complication concerns the ¸list). If this ¸list) contains the list item
“...”, this list item is replaced by the “missing values.” More precisely, the following happens:
Normally, when a list item ... is encountered, there should already have been two list items before it,
which where numbers. Examples of numbers are 1, -10, or -0.24. Let us call these numbers x and y
and let d := y − x be their difference. Next, there should also be one number following the three dots,
let us call this number z.
In this situation, the part of the list reading “x,y,...,z” is replaced by “x, x +d, x + 2d, x + 3d, . . . ,
x +md,” where the last dots are semantic dots, not syntactic dots. The value m is the largest number
such that x +md ≤ z if d is positive or such that x +md ≥ z if d is negative.
Perhaps it is best to explain this by some examples: The following ¸list) have the same effects:
\foreach \x in {1,2,...,6} {\x, } yields 1, 2, 3, 4, 5, 6,
246
\foreach \x in {1,2,3,...,6} {\x, } yields 1, 2, 3, 4, 5, 6,
\foreach \x in {1,3,...,11} {\x, } yields 1, 3, 5, 7, 9, 11,
\foreach \x in {1,3,...,10} {\x, } yields 1, 3, 5, 7, 9,
\foreach \x in {0,0.1,...,0.5} {\x, } yields 0, 0.1, 0.20001, 0.30002, 0.40002,
\foreach \x in {a,b,9,8,...,1,2,2.125,...,2.5} {\x, } yields a, b, 9, 8, 7, 6, 5, 4, 3, 2, 1, 2,
2.125, 2.25, 2.375, 2.5,
As can be seen, for fractional steps that are not multiples of 2
−n
for some small n, rounding errors
can occur pretty easily. Thus, in the second last case, 0.5 should probably be replaced by 0.501 for
robustness.
There is yet another special case for the ... statement: If the ... is used right after the first item in
the list, that is, if there is an x, but no y, the difference d obviously cannot be computed and is set to
1 if the number z following the dots is larger than x and is set to −1 if z is smaller:
\foreach \x in {1,...,6} {\x, } yields 1, 2, 3, 4, 5, 6,
\foreach \x in {9,...,3.5} {\x, } yields 9, 8, 7, 6, 5, 4,
Special handling of pairs. Different list items are separated by commas. However, this causes a
problem when the list items contain commas themselves as pairs like (0,1) do. In this case, you should
put the items containing commas in braces as in {(0,1)}. However, since pairs are such a natural and
useful case, they get a special treatment by the \foreach statement. When a list item starts with a (
everything up to the next ) is made part of the item. Thus, we can write things like the following:
\tikz
\foreach \position in {(0,0), (1,1), (2,0), (3,1)}
\draw \position rectangle +(.25,.5);
Using the foreach-statement inside paths. TikZ allows you to use a \foreach statement inside
a path construction. In such a case, the ¸commands) must be path construction commands. Here are
two examples:
\tikz
\draw (0,0)
\foreach \x in {1,...,3}
{ -- (\x,1) -- (\x,0) }
;
\tikz \draw \foreach \p in {1,...,3} {(\p,1)--(\p,3) (1,\p)--(3,\p)};
Multiple variables. You will often wish to iterate over two variables at the same time. Since you
can nest \foreach loops, this is normally straight-forward. However, you sometimes wish variables to
iterate “simultaneously.” For example, we might be given a list of edges that connect two coordinates
and might wish to iterate over these edges. While doing so, we would like the source and target of the
edges to be set to two different variables.
To achieve this, you can use the following syntax: The ¸variables) may not only be a single T
E
X-variable.
Instead, it can also be a list of variables separated by slashes (/). In this case the list items can also be
lists of values separated by slashes.
Assuming that the ¸variables) and the list items are lists of values, each time the ¸commands) are
executed, each of the variables in ¸variables) is set to one part of the list making up the current list
item. Here is an example to clarify this:
Example: \foreach \x / \y in {1/2,a/b} {‘‘\x\ and \y’’} yields “1 and 2”“a and b”.
If some entry in the ¸list) does not have “enough” slashes, the last entry will be repeated. Here is an
example:
247
0 1 2 3
e
\begin{tikzpicture}
\foreach \x/\xtext in {0,...,3,2.72 / e}
\draw (\x,0) node{$\xtext$};
\end{tikzpicture}
Here are more useful examples:
a
b
c
d
\begin{tikzpicture}
% Define some coordinates:
\tikzstyle{every node}=[draw,fill]
\path[shape=circle,fill=examplefill]
(0,0) node(a) {a}
(2,0.55) node(b) {b}
(1,1.5) node(c) {c}
(2,1.75) node(d) {d};
% Draw some connections:
\foreach \source/\target in {a/b, b/c, c/a, c/d}
\draw (\source) .. controls +(.75cm,0pt) and +(-.75cm,0pt)..(\target);
\end{tikzpicture}
\begin{tikzpicture}
% Let’s draw circles at interesting points:
\foreach \x / \y / \diameter in {0 / 0 / 2mm, 1 / 1 / 3mm, 2 / 0 / 1mm}
\draw (\x,\y) circle (\diameter);
% Same effect
\foreach \center/\diameter in {{(0,0)/2mm}, {(1,1)/3mm}, {(2,0)/1mm}}
\draw[yshift=2.5cm] \center circle (\diameter);
\end{tikzpicture}
3
2
1
12
11
10
9
8
7
6
5
4
\begin{tikzpicture}[cap=round,line width=3pt]
\filldraw [fill=examplefill] (0,0) circle (2cm);
\foreach \angle / \label in
{0/3, 30/2, 60/1, 90/12, 120/11, 150/10, 180/9,
210/8, 240/7, 270/6, 300/5, 330/4}
{
\draw[line width=1pt] (\angle:1.8cm) -- (\angle:2cm);
\draw (\angle:1.4cm) node{\textsf{\label}};
}
\foreach \angle in {0,90,180,270}
\draw[line width=2pt] (\angle:1.6cm) -- (\angle:2cm);
\draw (0,0) -- (120:0.8cm); % hour
\draw (0,0) -- (90:1cm); % minute
\end{tikzpicture}%
248
\tikz[shading=ball]
\foreach \x / \cola in {0/red,1/green,2/blue,3/yellow}
\foreach \y / \colb in {0/red,1/green,2/blue,3/yellow}
\shade[ball color=\cola!50!\colb] (\x,\y) circle (0.4cm);
\breakforeach
If this command is given inside a \foreach command, no further executions of the ¸commands) will
occur. However, the current execution of the ¸commands) is continued normally, so it is probably best
to use this command only at the end of a \foreach command.
\begin{tikzpicture}
\foreach \x in {1,...,4}
\foreach \y in {1,...,4}
{
\fill[red!50] (\x,\y) ellipse (3pt and 6pt);
\ifnum \x<\y
\breakforeach
\fi
}
\end{tikzpicture}
249
35 Date and Calendar Utility Macros
This section describes the package pgfcalendar.
\usepackage{pgfcalendar} % L
A
T
E
X
\input pgfcalendar.tex % plain T
E
X
\usemodule[pgfcalendar] % ConT
E
Xt
This package can be used independently of pgf. It has two purposes:
1. It provides functions for working with dates. Most noticably, it can convert a date in ISO-standard
format (like 1975-12-26) to a so-called Julian day number, which is defined in Wikipedia as follows:
“The Julian day or Julian day number is the (integer) number of days that have elapsed since
the initial epoch at noon Universal Time (UT) Monday, January 1, 4713 BC in the proleptic
Julian calendar.” The package also provides a function for converting a Julian day number to an
ISO-format date.
Julian day numbers make it very easy to work with days. For example, the date ten days in the
future of 2008-02-20 can be computed by converting this date to a Julian day number, adding 10,
and then converting it back. Also, the day of week of a given date can be computed by taking the
Julian day number modulo 7.
2. It provides a macro for typesetting a calendar. This macro is highly configurable and flexible (for
example, it can produce both plain text calendars and also complicated TikZ-based calendars), but
most users will not use the macro directly. It is the job of a frontend to provide useful configruations
for typesetting calendars based on this command.
35.1 Handling Dates
35.1.1 Conversions Between Date Types
\pgfcalendardatetojulian{¸date)}{¸counter)}
This macro converts a date in a format to be described in a moment to the Julian day number in the
Gregorian calendar. The ¸date) should expand to a string of the following form:
1. It should start with a number representing the year. Use \year for the current year, that is, the
year the file is being typeset.
2. The year must be followed by a hyphen.
3. Next should come a number representing the month. Use \month for the current month. You can,
but need not, use leading zeros. For example, 02 represents February, just like 2.
4. The month must also be followed by a hyphen.
5. Next you must either provide a day of month (again, a number and, again, \day yields the current
day of month) or the keyword last. This keyword refers to the last day of the month, which is
automatically computed (and which is a bit tricky to compute, especially for February).
6. Optionally, you can next provide a plus sign followed by positive or negative number. This number
of days will be added to the computed date.
Here are some examples:
• 2006-01-01 refers to the first day of 2006.
• 2006-02-last refers to February 28, 2006.
• \year-\month-\day refers to today.
• 2006-01-01+2 refser to January 3, 2006.
• \year-\month-\day+1 refers to tomorrow.
• \year-\month-\day+-1 refers to yesterday.
The conversion method is taken from the English Wikipedia entry on Julian days.
Example: \pgfcalendardatetojulian{2007-01-14}{\mycount} sets \mycount to 2454115.
250
\pgfcalendarjuliantodate{¸Julian day)}{¸year macro)}{¸month macro)}{¸day macro)}
This command converts a Julian day number to an ISO-date. The ¸Julian day) must be a number or
T
E
X counter, the ¸year macro), ¸month macro) and ¸day macro) must be T
E
X macro names. They
will be set to numbers representing the year, month, and day of the given Julian day in the Gregorian
calendar.
The ¸year macro) will be assigned the year without leading zeros. Note that this macro will produce
year 0 (as opposed to other calendars, where year 0 does not exist). However, if you really need calendars
for before the year 1, it is expected that you know what you are doing anyway.
The ¸month macro) gets assigned a two-digit number representing the month (with a leading zero, if
necessary). Thus, the macro is set to 01 for January.
The ¸day macro) gets assigned a two-digit number representing the day of the month (again with a
possible leading zero).
To convert a Julian day number to an ISO-date you use code like the following:
\pgfcalendardatetojulian{2454115}{\myyear}{\mymonth}{\myday}
\edef\isodate{\myyear-\mymonth-\myday}
The above code sets \isodate to 2007-01-14.
\pgfcalendarjuliantoweekday{¸Julian day)}{¸week day counter)}
This command converts a Julian day to a week day by computing the day modulo 7. The ¸week day
counter) must be a T
E
X counter. It will be set to 0 for a Monday, to 1 for a Tuesday, and so on.
Example: \pgfcalendarjuliantoweekday{2454115}{\mycount} sets \mycount to 6.
35.1.2 Checking Dates
\pgfcalendarifdate{¸date)}{¸tests)}{¸code)}{¸else code)}
This command is used to execute code based on properties of ¸date). The ¸date) must be a date in
ISO-format. For this date, the ¸tests) are checked (to be detailed later) and if one of the tests applied,
the ¸code) is executed. If none of the tests applies, the ¸else code) is executed.
Example: \pgfcalendarifdate{2007-02-07}{Wednesday}{Is a Wednesday}{Is not a Wednesday}
yields Is a Wednesday.
The ¸tests) is a comma-separated list of key-value pairs. The following are defined by default:
• all This test is passed by all dates.
• Monday This test is passed by all dates that are Mondays.
• Tuesday as above.
• Wednesday as above.
• Thursday as above.
• Friday as above.
• Saturday as above.
• Sunday as above.
• workday Passed by Mondays, Tuesdays, Wednesdays, Thursdays, and Fridays.
• weekend Passed Saturdays and Sundays.
• equals=¸reference) The ¸reference) can be in one of two forms: Either, it is a full ISO format date
like 2007-01-01 or the year may be missing as in 12-31. In the first case, the test is passed if
¸date) is the same as ¸reference). In the second case, the test is passed if the month and day part
of ¸date) is the same as ¸reference).
For example, the test equals=2007-01-10 will only be passed by this particular date. The test
equals=05-01 will be passed by every first of May on any year.
• at least=¸reference) This test works similarly to the equals test, only it is checked whether
¸date) is equal to ¸reference) or to any later date. Again, the ¸reference) can be a full date like
2007-01-01 or a short version like 07-01. For example, at least=07-01 is true for every day in
the second half of any year.
251
• at most=¸reference) as above.
• between=¸start reference) and ¸end refernce) This test checks whether the current date lies be-
tween the two given reference dates. Both full and short version may be given.
For example between=2007-01-01 and 2007-02-28 is true for the days in January and February
of 2007.
For another example, between=05-01 and 05-07 is true for the days of the first week of May of
any year.
• day of month=¸number) Passed by the day of month of the ¸date) is ¸number). For example, the
test day of month=1 is passed by every first of every month.
• end of month=¸number) Passed by the day of month of the ¸date) that is ¸number) from the end
of the month. For example, the test end of month=1 is passed by the last day of every month,
the test end of month=2 is passed by the second last day of every month. If ¸number) is omitted,
it is assumed to be 1.
In addition to the above checks, you can also define new checks. To do so, you must add a new key to
the key-value group pgfcalendar using \define@key. The job of the code of this new key is to possibly
set the T
E
X-if \ifpgfcalendarmatches to true (if it is already true, no action should be taken) to
indicate that the ¸date) passes the test setup by this new key.
In order to perform the test, the key code needs to know the date that should be checked. The date is
available through a macro, but a whole bunch of additional information about this date is also available
through the following macros:
• \pgfcalendarifdatejulian is the Julian day number of the ¸date) to be checked.
• \pgfcalendarifdateweekday is the weekday of the ¸date) to be checked.
• \pgfcalendarifdateyear is the year of the ¸date) to be checked.
• \pgfcalendarifdatemonth is the month of the ¸date) to be checked.
• \pgfcalendarifdateday is the day of month of the ¸date) to be checked.
For example, let us define a new key that checks whether the ¸date) is a Workers day (first of May).
This can be done as follows:
\define@key{pgfcalendar}{workers day}[]
{
\ifnum\pgfcalendarifdatemonth=5\relax
\ifnum\pgfcalendarifdateday=1\relax
\pgfcalendarmatchestrue
\fi
\fi
}
35.1.3 Typesetting Dates
\pgfcalendarweekdayname{¸week day number)}
This command expands to a textual representation of the day of week, given by the ¸week day number).
Thus, \pgfcalendarweekdayname{0} expands to Monday if the current language is English and to
Montag if the current language is German, and so on. See Section 35.1.4 for more details on translations.
Example: \pgfcalendarweekdayname{2} yields Wednesday.
\pgfcalendarweekdayshortname{¸week day number)}
This command works similiarly to the previous command, only an abbreviated version of the week day
is produced.
Example: \pgfcalendarweekdayshortname{2} yields Wed.
\pgfcalendarmonthname{¸month number)}
This command expands to a textual representation of the month, which is given by the ¸month number).
Example: \pgfcalendarmonthname{12} yields December.
252
\pgfcalendarmonthshortname{¸month number)}
As above, only an abbreviated version is produced.
Example: \pgfcalendarmonthshortname{12} yields Dec.
35.1.4 Localization
All textual representations of week days or months (like “Monday” or “February”) are wrapped with
\translate commands from the translator package (it this package is not loaded, no translation takes
place). Furthermore, the pgfcalendar package will try to load the translator-months-dictionary, if the
translator package is loaded.
The net effect of all this is that all dates will be translated to the current language setup in the translator
package. See the documentation of this package for more details.
35.2 Typesetting Calendars
\pgfcalendar{¸prefix)}{¸start date)}{¸end date)}{¸rendering code)}
This command can be used to typeset a calendar. It is a very general command, the actual work has to
be done by giving clever implementations of ¸rendering code). Note that this macro need not be called
inside a {pgfpicture} environment (even though it typically will be) and you can use it to typeset
calendars in normal T
E
X or using packages other than pgf.
Basic typesetting process. A calendar is typeset as follows: The ¸start date) and ¸end date) specify
a range of dates. For each date in this range the ¸rendering code) is executed with certain macros setup
to yield information about the current date (the current date in the enumeration of dates of the range).
Typically, the ¸rendering code) places nodes inside a picture, but it can do other things as well. Note
that it is also the job of the ¸rendering code) to position the calendar correctly.
The different calls of the ¸rending code) are not surrounded by T
E
X groups (though you can do so
yourself, of course). This means that settings can accumulate between different calls, which is often
desirable and useful.
Information about the current date. Inside the ¸rendering code), different macros can be access:
• \pgfcalendarprefix The ¸prefix) parameter. This prefix is recomended for nodes inside the
calendar, but you have to use it yourself explicitly.
• \pgfcalendarbeginiso The ¸start date) of range being typeset in ISO format (like 2006-01-10).
• \pgfcalendarbeginjulian Julian day number of ¸start date).
• \pgfcalendarendiso The ¸end date) of range being typeset in ISO format.
• \pgfcalendarendjulian Julian day number of ¸end date).
• \pgfcalendarcurrentjulian This T
E
X count holds the Julian day number of day currently begin
rendered.
• \pgfcalendarcurrentweekday The weekday (a number with zero representing Monday) of the
current date.
• \pgfcalendarcurrentyear The year of the current date.
• \pgfcalendarcurrentmonth The month of the current date (always two digits with a leading zero,
if necessary).
• \pgfcalendarcurrentday The day of month of the current date (alwyas two digits).
The \ifdate command. Inside the \pgfcalendar the macro \ifdate is available locally:
\ifdate{¸tests)}{¸code)}{¸else code)}
This command has the same effect as calling \pgfcalendarifdate for the current date.
Examples. In a first example, let us create a very simple calendar: It just lists the dates in a certain
range.
20 21 22 23 24 25 26 27 28 29 30 31 01 02 03 04 05 06 07 08 09 10
253
\pgfcalendar{cal}{2007-01-20}{2007-02-10}{\pgfcalendarcurrentday\ }
Let us now make this a little more interesting: Let us add a line break after each Sunday.
20 21
22 23 24 25 26 27 28
29 30 31 01 02 03 04
05 06 07 08 09 10
\pgfcalendar{cal}{2007-01-20}{2007-02-10}
{
\pgfcalendarcurrentday\
\ifdate{Sunday}{\par}{}
}
We now want to have all Mondays to be aligned on a column. For this, different approaches work. Here
is one based positioning each day horizontally using a skip.
20 21
22 23 24 25 26 27 28
29 30 31 01 02 03 04
05 06 07 08 09 10
\pgfcalendar{cal}{2007-01-20}{2007-02-10}
{%
\leavevmode%
\hbox to0pt{\hskip\pgfcalendarcurrentweekday cm\pgfcalendarcurrentday\hss}%
\ifdate{Sunday}{\par}{}%
}
Let us now typeset two complete months.
January
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31
February
1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28
\pgfcalendar{cal}{2007-01-01}{2007-02-28}{%
\ifdate{day of month=1}{
\par\bigskip\hbox to7.5cm{\itshape\hss\pgfcalendarshorthand mt\hss}\par
}{}%
\leavevmode%
{%
\ifdate{weekend}{\color{black!50}}{\color{black}}%
\hbox to0pt{%
\hskip\pgfcalendarcurrentweekday cm%
\hbox to1cm{\hss\pgfcalendarshorthand d-}\hss%
}%
}%
\ifdate{Sunday}{\par}{}%
}
For our final example, we use a {tikzpicture}.
254
20 21
22 23 24 25 26 27 28
29 30 31 01 02 03 04
05 06 07 08 09 10
\begin{tikzpicture}
\pgfcalendar{cal}{2007-01-20}{2007-02-10}{%
\ifdate{workday}{\tikzstyle{filling}=[fill=blue!20]}{\tikzstyle{filling}=[fill=red!20]}
\node (\pgfcalendarsuggestedname) at (\pgfcalendarcurrentweekday,0)
[anchor=base,circle,filling] {\pgfcalendarcurrentday};
\ifdate{Sunday}{\pgftransformyshift{-3em}}{}%
}
\draw (cal-2007-01-21) -- (cal-2007-02-03);
\end{tikzpicture}
\pgfcalendarshorthand{¸kind)}{¸representation)}
This command can be used inside a \pgfcalendar, where it will expand to a representation of the current
day, month, year or day of week, depending on whether ¸kind) is d, m, y or w. The ¸representation) can
be one of the following: -, =, 0, ., and t. They have the following meanings:
• The minus sign selects the shortest numerical representation possible (no leading zeros).
• The equal sign also selects the shortest numerical representation, but a space is added to single
digit days and months (thereby ensuring that they have the same length as other days).
• The zero digit selects a two-digit numerical representation for days and months. For years it is
allowed, but has no effect.
• The letter t selects a textual representation.
• The dot selects an abbreviated textual representation.
Normally, you should say \let\%=\pgfcalendarshorthand locally, so that you can write \%wt instead
of the much more cumbersome \pgfcalendarshorthand{w}{t}.
ISO form: 2007-01-20, long form: Saturday, January 20, 2007
\let\%=\pgfcalendarshorthand
\pgfcalendar{cal}{2007-01-20}{2007-01-20}
{ ISO form: \%y0-\%m0-\%d0, long form: \%wt, \%mt \%d-, \%y0}
\pgfcalendarsuggestedname
This macro expands to a suggested name for nodes representing days in a calendar. If the ¸prefix) is
empty, it expands to the empty string, otherwise it expands to the ¸prefix) of the calendar, followed by
a hyphen, followed by the ISO format version of the date. Thus, when the date 2007-01-01 is typeset
in a calendar for the prefix mycal, the macro expands to mycal-2007-01-01.
255
36 Page Management
This section describes the pgfpages packages. Although this package is not concerned with creating pictures,
its implementation relies so heavily on pgf that it is documented here. Currently, pgfpages only works
with L
A
T
E
X, but if you are adventurous, feel free to hack the code so that it also works with plain T
E
X.
The aim of pgfpages is to provide a flexible way of putting multiple pages on a single page inside T
E
X.
Thus, pgfpages is quite different from useful tools like psnup or pdfnup insofar as it creates its output in a
single pass. Furthermore, it works uniformly with both latex and pdflatex, making it easy to put multiple
pages on a single page without any fuss.
A word of warning: using pgfpages will destroy hyperlinks. Actually, the hyperlinks are not destroyed,
only they will appear at totally wrong positions on the final output. This is due to a fundamental flaw in
the pdf specification: In pdf the bounding rectangle of a hyperlink is given in “absolute page coordinates”
and translations or rotations do not affect them. Thus, the transformations applied by pgfpages to put the
pages where you want them are (cannot, even) be applied to the coordinates of hyperlinks. It is unlikely
that this will change in the foreseeable future.
36.1 Basic Usage
The internals of pgfpages are complex since the package can do all sorts of interesting tricks. For this
reason, so-called layouts are predefined that setup all option in appropriate ways.
You use a layout as follows:
\documentclass{article}
\usepackage{pgfpages}
\pgfpagesuselayout{2 on 1}[a4paper,landscape,border shrink=5mm]
\begin{document}
This text is shown on the left.
\clearpage
This text is shown on the right.
\end{document}
The layout 2 on 1 puts two pages on a single page. The option a4paper tells pgfpages that the resulting
page (called the physical page in the following) should be a4paper and it should be landscape (which is quite
logical since putting two portrait pages next to each other gives a landscape page). Normally, the logical
pages, that is, the pages that T
E
X “thinks” that it is typesetting, will have the same sizes, but this need not
be the case. pgfpages will automatically scale down the logical pages such that two logical pages fit next
to each other inside a DIN A4 page.
The border shrink tells pgfpages that it should add an additional 5mm to the shrinking such that a
5mm-wide border is shown around the resulting logical pages.
As a second example, let us put two pages produced by the beamer class on a single page:
\documentclass{beamer}
\usepackage{pgfpages}
\pgfpagesuselayout{2 on 1}[a4paper,border shrink=5mm]
\begin{document}
\begin{frame}
This text is shown at the top.
\end{frame}
\begin{frame}
This text is shown at the bottom.
\end{frame}
\end{document}
Note that we do not use the landscape option since beamer’s logical pages are already in landscape
mode and putting two landscape pages on top of each other results in a portrait page. However, if you had
used the 4 on 1 layout, you would have had to add landscape once more, using the 8 on 1 you must not,
using 16 on 1 you need it yet again. And, no, there is no 32 on 1 layout.
Another word of caution: using pgfpages will produce wrong page numbers in the .aux file. The rea-
son is that T
E
X instantiates the page numbers when writing an .aux file only when the physical page
is shipped out. Fortunately, this problem is easy to fix: First, typeset our file normally without using
the \pgfpagesuselayout command (just put the comment marker % before it) Then, rerun T
E
X with the
256
\pgfpagesuselayout command included and add the command \nofiles. This command ensures that the
.aux file is not modified, which is exactly what you want. So, to typeset the above example, you should
actually first T
E
X the following file:
\documentclass{article}
\usepackage{pgfpages}
%%\pgfpagesuselayout{2 on 1}[a4paper,landscape,border shrink=5mm]
%%\nofiles
\begin{document}
This text is shown on the left.
\clearpage
This text is shown on the right.
\end{document}
and then typeset
\documentclass{article}
\usepackage{pgfpages}
\pgfpagesuselayout{2 on 1}[a4paper,landscape,border shrink=5mm]
\nofiles
\begin{document}
This text is shown on the left.
\clearpage
This text is shown on the right.
\end{document}
The final basic example is the resize to layout (it works a bit like a hypothetical 1 on 1 layout). This
layout resizes the logical page such that is fits the specified physical size. Since this does not change the page
numbering, you need not worry about the .aux files with this layout. For example, adding the following
lines will ensure that the physical output will fit on DIN A4 paper:
\usepackage{pgfpages}
\pgfpagesuselayout{resize to}[a4paper]
This can be very useful when you have to handle lots of papers that are typeset for, say, letter paper and
you have an A4 printer or the other way round. For example, the following article will be fit for printing on
letter paper:
\documentclass[a4paper]{article}
%% a4 is currently the logical size and also the physical size
\usepackage{pgfpages}
\pgfpagesuselayout{resize to}[letterpaper]
%% a4 is still the logical size, but letter is the physical one
\begin{document}
\title{My Great Article}
...
\end{document}
36.2 The Predefined Layouts
This section explains the predefined layouts in more detail. You select a layout using the following command:
\pgfpagesuselayout{¸layout)}[¸options)]
Installs the specified ¸layout) with the given ¸options) set. The predefined layouts and their permissible
options are explained below.
If this function is called multiple times, only the last call “wins.” You can thereby overwrite any previous
settings. In particular, layouts do not accumulate.
Example: \pgfpagesuselayout{resize to}[a4paper]
\pgfpagesuselayout{resize to}[¸options)]
This layout is used to resize every logical page to a specified physical size. To determine the target size,
the following options may be given:
257
• physical paper height=¸size) sets the height of the physical pape size to ¸size).
• physical paper width=¸size) sets the width of the physical pape size to ¸size).
• a0paper sets the physical page size to DIN A0 paper.
• a1paper sets the physical page size to DIN A1 paper.
• a2paper sets the physical page size to DIN A2 paper.
• a3paper sets the physical page size to DIN A3 paper.
• a4paper sets the physical page size to DIN A4 paper.
• a5paper sets the physical page size to DIN A5 paper.
• a6paper sets the physical page size to DIN A6 paper.
• letterpaper sets the physical page size to the American letter paper size.
• legalpaper sets the physical page size to the American legal paper size.
• executivepaper sets the physical page size to the American executive paper size.
• landscape swaps the height and the width of the physical paper.
• border shrink=¸size) additionally reduces the size of the logical page on the physical page by
¸size).
\pgfpagesuselayout{2 on 1}[¸options)]
Puts two logical pages alongside each other on each physical page if the logical height is larger than the
logical width (logical pages are in portrait mode). Otherwise, two logical pages are put on top of each
other (logical pages are in landscape mode). When using this layout, it is advisable to use the \nofiles
command, but this is not done automatically.
The same ¸options) as for the resize to layout an be used, plus the following option:
• odd numbered pages right places the first page on the right.
\pgfpagesuselayout{4 on 1}[¸options)]
Puts four logical pages on a single physical page. The same ¸options) as for the resize to layout an
be used.
\pgfpagesuselayout{8 on 1}[¸options)]
Puts eight logical pages on a single physical page. As for 2 on 1, the orientation depends on whether
the logical pages are in landscape mode or in portrait mode.
\pgfpagesuselayout{16 on 1}[¸options)]
This is for the ceo.
\pgfpagesuselayout{rounded corners}[¸options)]
This layout adds “rounded corners” to every page, which, supposedly, looks nicer during presentations
with projectors (personally, I doubt this). This is done by (possibly) resizing the page to the physical
page size. Then four black rectangles are drawn in each corner. Next, a clipping region is set up that
contains all of the logical page except for little rounded corners. Finally, the logical page is draw, clipped
against the clipping region.
Note that every logical page should fill its background for this to work.
In addition to the ¸options) that can be given to resize to the following options may be given.
• corner width=¸size) specifies the size of the corner.
\documentclass{beamer}
\usepackage{pgfpages}
\pgfpagesuselayout{rounded corners}[corner width=5pt]
\begin{document}
...
\end{document}
258
\pgfpagesuselayout{two screens with lagging second}[¸options)]
This layout puts two logical pages alongside each other. The second page always shows what the main
page showed on the previous physical page. Thus, the second page “lags behind” the main page. This
can be useful when you have to projectors attached to your computer and can show different parts of a
physical page on different projectors.
The following ¸options) may be given:
• second right puts the second page right of the main page. This will make the physical pages
twice as wide as the logical pages, but it will retain the height.
• second left puts the second page left, otherwise it behave the same as second right.
• second bottom puts the second page below the main page. This make the physical pages twice as
high as the logical ones.
• second top works like second bottom.
\pgfpagesuselayout{two screens with optional second}[¸options)]
This layout works similarly to two screens with lagging second. The difference is that the contents
of the second screen only changes when one of the commands \pgfshipoutlogicalpage{2}{¸box)} or
\pgfcurrentpagewillbelogicalpage{2} is called. The first puts the given ¸box) on the second page.
The second specifies that the current page should be put there, once it is finished.
The same options as for two screens with lagging second may be given.
You can define your own predefined layouts using the following command:
\pgfpagesdeclarelayout{¸layout)}{¸before actions)}{¸after actions)}
This command predefines a ¸layout) that can later be installed using the \pgfpagesuselayout com-
mand.
When \pgfpagesuselayout{¸layout)}[¸options)] is called, the following happens: First, the ¸before
actions) are executed. They can be used, for example, to setup default values for keys. Next,
\setkeys{pgfpagesuselayoutoption}{¸options)} is executed. Finally, the ¸after actions) are exe-
cuted.
Here is an example:
\pgfpagesdeclarelayout{resize to}
{
\def\pgfpageoptionborder{0pt}
}
{
\pgfpagesphysicalpageoptions
{%
logical pages=1,%
physical height=\pgfpageoptionheight,%
physical width=\pgfpageoptionwidth%
}
\pgfpageslogicalpageoptions{1}
{%
resized width=\pgfphysicalwidth,%
resized height=\pgfphysicalheight,%
border shrink=\pgfpageoptionborder,%
center=\pgfpoint{.5\pgfphysicalwidth}{.5\pgfphysicalheight}%
}%
}
36.3 Defining a Layout
If none of the predefined layouts meets your problem or if you wish to modify them, you can create layouts
from scratch. This section explains how this is done.
Basically, pgfpages hooks into T
E
X’s \shipout function. This function is called whenever T
E
X has
completed typesetting a page and wishes to send this page to the .dvi or .pdf file. The pgfpages package
redefines this command. Instead of sending the page to the output file, pgfpages stores it in an internal box
and then acts as if the page had been output. When T
E
X tries to output the next page using \shipout, this
call is once more intercepted and the page is stored in another box. These boxes are called logical pages.
259
At some point, enough logical pages have been accumulated such that a physical page can be output.
When this happens, pgfpages possibly scales, rotates, and translates the logical pages (and possibly even
does further modifications) and then puts them at certain positions of the physical page. Once this page is
fully assembled, the “real” or “original” \shipout is called to send the physical page to the output file.
In reality, things are slightly more complicated. First, once a physical page has been shipped out, the
logical pages are usually voided, but this need not be the case. Instead, it is possible that certain logical
page just retain their contents after the physical page has been shipped out and these pages need not be
filled once more before a physical shipout can occur. However, the contents of these logical pages can still
be changed using special commands. It is also possible that after a shipout certain logical pages are filled
with the contents of other logical pages.
A layout defines for each logical page where it will go on the physical page and which further modifications
should be done. The following two commands are used to define the layout:
\pgfpagesphysicalpageoptions{¸options)}
This command sets the characteristic of the “physical” page. For example, it is used to specify
how many logical pages there are and how many logical pages must be accumulated before a physi-
cal page is shipped out. How each individual logical page is typeset is specified using the command
\pgfpageslogicalpageoptions, described later.
Example: A layout for putting two portrait pages on a single landscape page:
\pgfpagesphysicalpageoptions
{%
logical pages=2,%
physical height=\paperwidth,%
physical width=\paperheight,%
}
\pgfpageslogicalpageoptions{1}
{%
resized width=.5\pgfphysicalwidth,%
resized height=\pgfphysicalheight,%
center=\pgfpoint{.25\pgfphysicalwidth}{.5\pgfphysicalheight}%
}%
\pgfpageslogicalpageoptions{2}
{%
resized width=.5\pgfphysicalwidth,%
resized height=\pgfphysicalheight,%
center=\pgfpoint{.75\pgfphysicalwidth}{.5\pgfphysicalheight}%
}%
The following ¸options) may be set:
• logical pages=¸logical pages) specified how many logical pages there are, in total. These are
numbered 1 to ¸logical pages).
• first logical shipout=¸first). See the the next option. By default, ¸first) is 1.
• last logical shipout=¸last). Together with the previous option, these two options define an
interval of pages inside the range 1 to ¸logical pages). Only this range is used to store the pages
that are shipped out by T
E
X. This means that after a physical shipout has just occured (or at the
beginning), the first time T
E
X wishes to perform a shipout, the page to be shipped out is stored
in logical page ¸first). The next time T
E
X performs a shipout, the page is stored in logical page
¸first) + 1 and so on, until the logical page ¸last) is also filled. Once this happens, a physical
shipout occurs and the process starts once more.
Note that logical pages that lie outside the interval between ¸first) and ¸last) are filled only indi-
rectly or when special commands are used.
By default, ¸last) equals ¸logical pages).
• current logical shipout=¸current) changes an internal counter such that T
E
X’s next logical
shipout will be stored in logical page ¸current).
This option can be used to “warp” the logical page filling mechanism to a certain page. You can
both skip logical pages and overwrite already filled logical pages. After the logical page ¸current)
has been filled, the internal counter is incremented normally as if the logical page ¸current) had
been “reached” normally. If you specify a ¸current) larger to ¸last), a physical shipout will occur
after the logical page ¸current) has been filled.
260
• physical height=¸height) specifies the height of the physical pages. This height is typically
different from the normal \paperheight, which is used by T
E
X for its typesetting and page breaking
purposes.
• physical width=¸width) specifies the physical width.
\pgfpageslogicalpageoptions{¸logical page number)}{¸options)}
This command is used to specify where the logical page number ¸logical page number) will be placed on
the physical page. In addition, this command can be used to install additional “code” to be executed
when this page is put on the physical page.
The number ¸logical page number) should be between 1 and ¸logical pages), which has previously been
installed using the \pgfpagesphysicalpageoptions command.
The following ¸options) may be given:
• center=¸pgf point) specifies the center of the logical page inside the physical page as a pgf-point.
The origin of the coordinate system of the physical page is at the lower left corner.
\pgfpageslogicalpageoptions{1}
{% center logical page on middle of left side
center=\pgfpoint{.25\pgfphysicalwidth}{.5\pgfphysicalheight}%
resized width=.5\pgfphysicalwidth,%
resized height=\pgfphysicalheight,%
}
• resized width=¸size) specifies the width that the logical page should have at most on the physical
page. To achieve this width, the pages is scaled down appropriately or more. The “or more” part
can happen if the resize height option is also used. In this case, the scaling is chosen such that
both the specified height and width are met. The aspect ratio of a logical page is not modified.
• resized height=¸height) specifies the maximum height of the logical page.
• original width=¸width) specifies the width the T
E
X “thinks” that the logical page has. This
width is \paperwidth at the point of invocation, by default. Note that setting this width to
something different from \paperwidth does not change the \pagewidth during T
E
X’s typesetting.
You have to do that yourself.
You need this option only for special logical pages that have a height or width different from the
normal one and for which you will (later on) set these sizes yourself.
• original height=¸height) works like original width.
• scale=¸factor) scales the page by at least the given ¸factor). A ¸factor) of 0.5 will half the size of
the page, a factor or 2 will double the size. “At least” means that if options like resize height
are given and if the scaling required to meet that option is less than ¸factor), that other scaling is
used instead.
• xscale=¸factor) scales the logical page along the x-axis by the given ¸factor). This scaling is done
independently of any other scaling. Mostly, this option is useful for a factor of -1, which flips the
page along the y-axis. The aspect ratio is not kept.
• yscale=¸factor) works like xscale, only for the y-axis.
• rotation=¸degree) rotates the page by ¸degree) around its center. Use a degree of 90 or -90 to go
from portrait to landscape and back. The rotation need not be a multiple of 90.
• copy from=¸logical page number). Normally, after a physical shipout has occured, all logical pages
are voided in a loop. However, if this option is given, the current logical page is filled with the
contents of the old logical page number ¸logical page number).
Example: Have logical page 2 retain its contents:
\pgfpageslogicalpageoptions{2}{copy from=2}
Example: Let logical page 2 show what logical page 1 showed on the just-shipped-out physical
page:
\pgfpageslogicalpageoptions{2}{copy from=1}
261
• border shrink=¸size) specifies an addition reduction of the size to which the page is page is scaled
down.
• border code=¸code). When this option is given, the ¸code) is executed before the page box
is inserted with a path preinstalled that is a rectangle around the current logical page. Thus,
setting ¸code) to \pgfstroke draws a rectangle around the logical page. Setting ¸code) to
\pgfsetlinewidth{3pt}\pgfstroke results in a thick (ugly) frame. Adding dashes and filling
can result in arbitrarily funky and distracting borders.
You can also call \pgfdiscardpath and add your own path construction code (for example to paint
a rectangle with rounded corners). The coordinate system is setup in such a way that a rectangle
starting at the origin and having the height and width of T
E
X-box 0 will result in a rectangle filling
exactly the logical page currently being put on the physical page. The logical page is inserted after
these commands have been executed.
Example: Add a rectangle around the page:
\pgfpageslogicalpageoptions{1}{border code=\pgfstroke}
• corner width=¸size) adds black “rounded corners” to the page. See the description of the prede-
fined layout rounded corners on page 258.
36.4 Creating Logical Pages
Logical pages are created whenever a T
E
X thinks that a page is full and performs a \shipout command.
This will cause pgfpages to store the box that was supposed to be shipped out internally until enough
logical pages have been collected such that a physical shipout can occur.
Normally, whenever a logical shipout occurs that current page is stored in logical page number ¸current
logical page). This counter is then incremented, until it is larger than ¸last logical shipout). You can, however,
directly change the value of ¸current logical page) by calling \pgfpagesphysicalpageoptions.
Another way to set the contents of a logical page is to use the following command:
\pgfpagesshipoutlogicalpage{¸number)}¸box)
This command sets to logical page ¸number) to ¸box). The ¸box) should be the code of a T
E
X box
command. This command does not influence the counter ¸current logical page) and does not cause a
physical shipout.
\pgfpagesshipoutlogicalpage{0}\vbox{Hi!}
This command can be used to set the contents of logical pages that are normally not filled.
The final way of setting a logical page is using the following command:
\pgfpagescurrentpagewillbelogicalpage{¸number)}
When the current T
E
X page has been typeset, it will be become the given logical page ¸number). This
command “interrupts” the normal order of logical pages, that is, it behaves like the previous command
and does not update the ¸current logical page) counter.
\pgfpagesuselayout{two screens with optional second}
...
Text for main page.
\clearpage
\pgfpagescurrentpagewillbelogicalpage{2}
Text that goes to second page
\clearpage
Text for main page.
262
37 Extended Color Support
This section documents the package xxcolor, which is currently distributed as part of pgf. This package
extends the xcolor package, written by Uwe Kern, which in turn extends the color package. I hope that
the commands in xxcolor will some day migrate to xcolor, such that this package becomes superfluous.
The main aim of the xxcolor package is to provide an environment inside which all colors are “washed
out” or “dimmed.” This is useful in numerous situations and must typically be achieved in a roundabout
manner if such an environment is not available.
\begin{colormixin}{¸mix-in specification)}
¸environment contents)
\end{colormixin}
The mix-in specification is applied to all colors inside the environment. At the beginning of the environ-
ment, the mix-in is applied to the current color, i. e., the color that was in effect before the environment
started. A mix-in specification is a number between 0 and 100 followed by an exclamation mark and a
color name. When a \color command is encountered inside a mix-in environment, the number states
what percentage of the desired color should be used. The rest is “filled up” with the color given in the
mix-in specification. Thus, a mix-in specification like 90!blue will mix in 10% of blue into everything,
whereas 25!white will make everything nearly white.
Red text,washed-out
red text, washed-out
blue text, dark
washed-out blue text,
dark washed-out green
text, back to
washed-out blue
text,and back to red.
\begin{minipage}{3.5cm}\raggedright
\color{red}Red text,%
\begin{colormixin}{25!white}
washed-out red text,
\color{blue} washed-out blue text,
\begin{colormixin}{25!black}
dark washed-out blue text,
\color{green} dark washed-out green text,%
\end{colormixin}
back to washed-out blue text,%
\end{colormixin}
and back to red.
\end{minipage}%
Note that the environment only changes colors that have been installed using the standard L
A
T
E
X \color
command. In particular, the colors in images are not changed. There is, however, some support offered by
the commands \pgfuseimage and \pgfuseshading. If the first command is invoked inside a colormixin
environment with the parameter, say, 50!black on an image with the name foo, the command will first check
whether there is also a defined image with the name foo.!50!black. If so, this image is used instead. This
allows you to provide a different image for this case. If you nest colormixin environments, the different mix-
ins are all appended. For example, inside the inner environment of the above example, \pgfuseimage{foo}
would first check whether there exists an image named foo.!50!white!25!black.
\colorcurrentmixin
Expands to the current accumulated mix-in. Each nesting of a colormixin adds a mix-in to this list.
!75!white should be “!75!white”
!75!black!75!white should be “!75!black!75!white”
!50!white!75!black!75!white should be “!50!white!75!black!75!white”
\begin{minipage}{\linewidth-6pt}\raggedright
\begin{colormixin}{75!white}
\colorcurrentmixin\ should be ‘‘!75!white’’\par
\begin{colormixin}{75!black}
\colorcurrentmixin\ should be ‘‘!75!black!75!white’’\par
\begin{colormixin}{50!white}
\colorcurrentmixin\ should be ‘‘!50!white!75!black!75!white’’\par
\end{colormixin}
\end{colormixin}
\end{colormixin}
\end{minipage}
263
Part VI
Mathematical Engine
by Mark Wibrow and Till Tantau
pgf comes with its own mathematical engine. The job of this engine is to support mathematical opera-
tions like addition, subtraction, multiplication and division, using both integers and non-integers, but also
functions such as square-roots, sine, cosine, and generate pseudo-random numbers.
Mostly, you will use the mathematical facilities of pgf indirectly, namely when you write a coordinate
like (5cm*3,6cm/4), but the mathematical engine can also be used independently of pgf and TikZ.
\pgfmathsetseed{1}
\foreach \col in {black,red,green,blue}
{
\begin{tikzpicture}[x=10pt,y=10pt,ultra thick,baseline,cap=round]
\coordinate (current point) at (0,0);
\coordinate (old velocity) at (0,0);
\coordinate (new velocity) at (rand,rand);
\foreach \i in {0,1,...,100}
{
\draw[\col!\i] (current point)
.. controls ++([scale=-1]old velocity) and
++(new velocity) .. ++(rand,rand)
coordinate (current point);
\coordinate (old velocity) at (new velocity);
\coordinate (new velocity) at (rand,rand);
}
\end{tikzpicture}
}
264
38 Design Principles
pgf needs to perform many computations while typesetting a picture. For this, pgf relies on a mathematical
engine, which can also be used independently of pgf, but which is distributed as part of the pgf package
nevertheless. Basically, the engine provides a parsing mechanism similar to the calc package so that
expressions like 2*3cm+5cm can be parsed; but the pgf engine is more powerful and can be extended and
enhanced.
pgf provides enhanced functionality, which permits the parsing of mathematical operations involving
integers and non-integers with or without units. Furthermore, various functions, including trigonometric
functions and random number generators can also be parsed (see Section 39.1). The calc macros \setlength
and friends have pgf versions which can parse these operations and functions (see Section 39.1). Additionally,
each operation and function has an independent pgf command associated with it (see Section 40), and can
be accessed outside the parser.
The mathematical engine of pgf is implicitly used whenever you specify a number or dimension in a
higher-level macro. For instance, you can write \pgfpoint{2cm+4cm/2}{3cm*sin(30)} or suchlike. How-
ever, the mathematical engine can also be used independently of the pgf core, that is, you can also just load
it to get access to a mathematical parser.
38.1 Loading the Mathematical Engine
The mathematical engine of pgf is loaded automatically by pgf, but if you wish to use the mathematical
engine but you do not need pgf itself, you can load the following package:
\usepackage{pgfmath} % L
A
T
E
X
\input pgfmath.tex % plain T
E
X
\usemodule[pgfmath] % ConT
E
Xt
This command will load the mathematical engine of pgf, but not pgf itself. It defines commands like
\pgfmathparse.
38.2 Layers of the Mathematical Engine
Like pgf itself, the mathematical engine is also structured into different layers:
1. The top layer, which you will typically use directly, provides the command \pgfmathparse. This
command parses a mathematical expression and evaluates it.
Additionally, the top layer also defines some additional functions similar to the macros of the calc pack-
age for setting dimensions and counters. These macros are just wrappers around the \pgfmathparse
macro.
2. The calculation layer provides macros for performing one specific computation like computing a recip-
rocal or a multiplication. The parser uses these macros for the actual computation.
3. The implementation layer provides the actual implementations of the computations. These can be
changed (and possibly be made more efficient) without affecting the higher layers.
38.3 Efficiency and Accuracy of the Mathematical Engine
Currently, the mathematical algorithms are all implemented in T
E
X. This poses some intriguing program-
ming challenges as T
E
X is a typesetting language not a mathematical one, and as with any programming
language, there is a trade-off between accuracy and efficiency. If you find the level of accuracy insufficient
for you purposes, you will have to replace the algorithms in the implementation layer.
All the fancy mathematical “bells-and-whistles” that the parser provides, come with an additional pro-
cessing cost, and in some instances, such as simply setting a length to 1cm, with no other operations involved,
the addition processing time is undesirable. To overcome this, the following feature is implemented: when
no mathematical operations are required, the value in ¸expression) can be preceded by +. This will bypass
the parsing process and the assignment will be orders of magnitude faster. This feature only works with the
macros for setting registers described in Section 39.1.
\pgfmathsetlength\mydimen{1cm} % parsed : slower.
\pgfmathsetlength\mydimen{+1cm} % not parsed : much faster.
265
39 Evaluating Mathematical Expressions
The easiest way of using pgf’s mathematical engine is to provide a mathematical expression given in the
usual infix notation (such as 1cm+4*2cm/5.5 or 2*3+3*sin(30)). This expression can be parsed by the
mathematical engine and the result be placed in a dimension register, a counter, or a macro. Supported are
infix mathematical operations involving integers and non-integers, with or without units.
It should be noted that all calculations must not exceed ±16383.99999 at any point, because the underly-
ing algorithms relie on T
E
X dimensions. This means that many of the underlying algorithms are necessarily
approximate. It also means that some of the algorithms are not very fast. T
E
X is, after all, a typesetting
language and not ideally suited to relatively advanced mathematical operations. However, it is possible to
change the algorithms as described in Section 41.
In the present section, the high-level macros for parsing an expression are explained first, then the syntax
for expression is explained.
39.1 Commands for Parsing Expressions
The basic command for invoking the parser of pgf’s mathematical engine is the following:
\pgfmathparse{¸expression)}
This macro parses ¸expression) and returns the result without units in the macro \pgfmathresult.
Example: \pgfmathparse{2pt+3.5pt} will set \pgfmathresult to the text 5.5.
In the following, the special properties of this command are explained. The exact syntax of mathematical
expressions is explained in Section 39.2. Note that unlike the rest of the manual, the examples show
the result of the calculation (that is, the value of the macro \pgfmathresult), not what is displayed in
the document.
• The result stored in the macro \pgfmathresult is a decimal without units. This is true regardless of
whether the ¸expression) contains any unit specification. But, any units specified will be converted
to points first.
5.4 \pgfmathparse{2pt+3.4pt}
153.64468 \pgfmathparse{2cm+3.4cm}
• If no units are specified at any point in the expression, the result will be multiplied by the value
in \pgfmathresultunitscale, which can be a number or a dimension (which will be converted to
points). By default it is set to 1, but can be changed with \pgfmathsetresultunitscale. Note
that the result will still be a number without units.
5.4 \pgfmathparse{2pt+3.4pt}
153.64464 \pgfmathsetresultunitscale{1cm}
\pgfmathparse{2+3.4}
• You can check whether an expression contained a unit using the T
E
X-if \ifpgfmathunitsdeclared.
After a call of \pgfmathparse this if will be true exactly if some unit was encountered in the
expression.
• The parser handles numbers with or without units regardless of the operation.
1.32544 \pgfmathparse{54pt/3cm*2.1}
• the parser can cope with T
E
X registers, including those preceded by \the.
42.34 \pgf@x=12.34pt
\c@pgf@counta=5
\pgfmathparse{\pgf@x+\c@pgf@counta*6}
113.56 \pgf@x=56.78pt
\pgfmathparse{\pgf@x+\the\pgf@x}
266
• T
E
X dimension registers can be multiplied without the * operator by preceding them with a number
(not a function), or a count register.
45.0 \c@pgf@counta=-4
\pgf@x=10pt
\pgfmathparse{.5\pgf@x-\c@pgf@counta\pgf@x}%
• Parenthesis can be used to group operations.
13.5 \pgfmathparse{(4pt+0.5)*3}
• functions are recognized, so it is possible to parse sin(.5*pi r)*60, which means “the sine of 0.5
times π radians, multiplied by 60”. The argument of most functions can be any expression.
59.99908 \pgfmathparse{sin(pi/2 r)*60}
• Scientific notation in the form 1.234e+4 is recognised (but the restriction on the range of values
still applies). The exponent symbol can be upper or lower case (i.e., E or e).
0.01234 \pgfmathparse{1.234567891e-2}
12345.67891 \pgfmathparse{1.234567891e4}
\pgfmathqparse{¸expression)}
This macro is similar to \pgfmathparse: it parses ¸expression) and returns the result in the macro
\pgfmathresult. It differs in two respects. Firstly, \pgfmathqparse does not parse functions or
scientific notation. Secondly, numbers in ¸expression) must specify a T
E
X unit (except in such instances
as 0.5\pgf@x), which greatly simplifies the problem of parsing of non-integers. As a result of these
restrictions \pgfmathqparse is about twice as fast as \pgfmathparse. Note that the result will still be
a number without units.
\pgfmathsetresultunitscale{¸number or dimension)}
Sets the value in \pgfmathresultunitscale, which scales the result of an expression parsed with
\pgfmathparse, if that expression contains no units at any point. The argument can be an integer,
non-integer or a dimension, but the result will still be a number without units. Note, that this will affect
\pgfmathsetlength and friends, but not if the expression starts with + (which switches parsing off).
By default the value in \pgfmathresultunitscale is 1.
Instead of the \pgfmathparse macro you can also wrapper commands, whose usage is very similar to
their cousins in the calc package. The only difference is that the expressions can be any expression that is
handled by \pgfmathparse.
For all of the following commands, if ¸expression) starts with +, no parsing is done and a simple assignment
or increment is done using normal T
E
X assignments or increments. This will be orders of magnitude faster
than calling the parser.
\pgfmathsetlength{¸dimension register)}{¸expression)}
Sets the length of the T
E
X ¸dimension register), to the value (in points) specified by ¸expression). The
¸expression) will be parsed using \pgfmathparse.
\pgfmathaddtolength{¸dimension register)}{¸expression)}
Adds the value (in points) of ¸expression) to the T
E
X ¸dimension register).
\pgfmathsetcount{¸count register)}{¸expression)}
Sets the value of the T
E
X ¸count register), to the truncated value specified by ¸expression).
\pgfmathaddtocount{¸count register)}{¸expression)}
Adds the truncated value of ¸expression) to the T
E
X ¸count register).
\pgfmathsetcounter{¸counter)}{¸expression)}
Sets the value of the ¸counter), to the truncated value specified by ¸expression).
267
\pgfmathaddtocounter{¸counter)}{¸expression)}
Adds the truncated value of ¸expression) to ¸counter).
\pgfmathsetmacro{¸macro)}{¸expression)}
Defines ¸macro) as the value of ¸expression). The result is a decimal without units.
39.2 Syntax for mathematical expressions
The syntax for the expressions recognized by \pgfmathparse and friends is based on the syntax recognized
by matlab. The following operations and functions are currently recognized:
x + y
Adds y to x.
6.0 \pgfmathparse{4+2pt}
x - y
Subtracts y from x.
41.53899 \pgfmathparse{155.35-4cm}
x * y
Multiplies x by y.
17.78395 \pgfmathparse{3.9pt*4.56}
x / y
Divides x by y.
-1.85881 \pgfmathparse{-31.6pt/17}
x ^ y
Raises x to the power y. y should be an integer, but it can be negative.
27.98418 \pgfmathparse{2.3^4}
0.0625 \pgfmathparse{2^-4}
x == y
This evaluates to 1 if x equals y, or 0 if x does not equal y. Note that equalities (and inequalities) are
evaluated left to right, and are only evaluated when another equality (or inequality) operator is scanned,
or the end of the current group or parse is reached. So 5+4==3+2==9 results in 0 because 5+4 does not
equal 3+2, resulting in zero, and the second equality is therefore evaluating 0==9.
1.0 \pgfmathparse{3*5 == 15}
x > y
This evaluates to 1 if x is greater than y, or 0 if x is smaller or equal to y.
1.0 \pgfmathparse{17 > 4.2*1.97+4}
x < y
This evaluates to 1 if x is smaller than y, or 0 if x is greater or equal to y.
0.0 \pgfmathparse{2 < -5.2/-3.6-2}
mod(x,y)
This evaluates x modulo y. This function cannot be nested inside itself or the functions max, min or
veclen.
2.0 \pgfmathparse{mod(20,6)}
max(x,y)
This evaluates to the maximum of x or y. This function cannot be nested inside itself or the functions
min, mod or veclen.
23.0 \pgfmathparse{max(17,23)}
268
min(x,y)
This evaluates to the minimum of x or y. This function cannot be nested inside itself or the functions
max, mod or veclen.
17.0 \pgfmathparse{min(17,23)}
abs(x)
Evaluates the absolute value of x.
5.0 \pgfmathparse{abs(-5)}
-12.0 \pgfmathparse{-abs(4*-3)}
round(x)
Rounds x to the nearest integer. It uses “asymmetric half-up” rounding. So 1.5 is rounded to 2, but
-1.5 is rounded to -2 (not 0).
2.0 \pgfmathparse{round(32.5/17)}
33.0 \pgfmathparse{round(398/12)}
floor(x)
Rounds x down to the nearest integer.
1.0 \pgfmathparse{floor(32.5/17)}
33.0 \pgfmathparse{floor(398/12)}
ceil(x)
Rounds x up to the nearest integer.
2.0 \pgfmathparse{ceil(32.5/17)}
34.0 \pgfmathparse{ceil(398/12)}
exp(x)
Maclaurin series for e
x
.
2.7182 \pgfmathparse{exp(1)}
10.3806 \pgfmathparse{exp(2.34)}
pow(x,y)
Raises x to the power y. y should be an integer, but it can be negative.
128.0 \pgfmathparse{pow(2,7)}
sqrt(x)
Calculates

x.
3.16228 \pgfmathparse{sqrt(10)}
93.62389 \pgfmathparse{sqrt(8765.432)}
veclen(x,y)
Evaluates the Euclidean distance from (0,0) to (x,y). This function cannot be nested inside itself, or
the functions max, min or mod.
20.51811 \pgfmathparse{veclen(15,14)}
4.99994 \pgfmathparse{veclen(3,4)}
pi
The constant π = 3.14159.
3.14159 \pgfmathparse{pi}
179.99962 \pgfmathparse{pi r}
269
x r
This converts x from radians to degrees. Note that r will evaluate any preceding series of multiplication
or division before conversion, but not other operations. So 3*4/6r converts 2 radians to degrees, but
3-4+6r, converts 6 radians to degrees and adds the result to -1.
179.99963 \pgfmathparse{2*pi r-pi r}
44.99924 \pgfmathparse{2*pi/8 r}
-59.99908 \pgfmathparse{sin(3*pi/2r)*60}
rad(x)
Convert x to radians. x is assumed to be in degrees.
1.57079 \pgfmathparse{rad(90)}
deg(x)
Convert x to degrees. x is assumed to be in radians.
269.999 \pgfmathparse{deg(3*pi/2)}
sin(x)
Sine of x. By employing the r operator, x can be in radians.
0.86603 \pgfmathparse{sin(60)}
0.86601 \pgfmathparse{sin(pi/3 r)}
cos(x)
Cosine of x. By employing the r operator, x can be in radians.
0.5 \pgfmathparse{cos(60)}
0.49998 \pgfmathparse{cos(pi/3 r)}
tan(x)
Tangent of x. By employing the r operator, x can be in radians.
1.0 \pgfmathparse{tan(45)}
1.0 \pgfmathparse{tan(2*pi/8 r)}
asin(x)
Arcsine of x. The result is in degrees.
44.99133 \pgfmathparse{asin(0.7071)}
acos(x)
Arccosine of x in degrees.
60.0 \pgfmathparse{acos(0.5)}
atan(x)
Arctangent of x in degrees.
45.0 \pgfmathparse{atan(1)}
rnd
Generates a pseudo-random number between 0 and 1.
0.32524 \pgfmathparse{rnd}
0.63 \pgfmathparse{2*rnd}
4.02074 \pgfmathparse{-rnd+5}
rand
Generates a pseudo-random number between -1 and 1.
0.80753 \pgfmathparse{rand}
8.9788 \pgfmathparse{rand*15}
270
40 Evaluating Mathematical Operations
Instead of parsing and evaluating complex expressions, you can also use the mathematical engine to evaluate
a single mathematical operation. The macros used for these computations are described in the following.
40.1 Basic Operations and Functions
\pgfmathadd{¸x)}{¸y)}
Defines \pgfmathresult as ¸x) + ¸y).
\pgfmathsubtract{¸x)}{¸y)}
Defines \pgfmathresult as ¸x) − ¸y).
\pgfmathmultiply{¸x)}{¸y)}
Defines \pgfmathresult as ¸x) ¸y).
\pgfmathdivide{¸x)}{¸y)}
Defines \pgfmathresult as ¸x) ÷ ¸y).
\pgfmathgreaterthan{¸x)}{¸y)}
Defines \pgfmathresult as 1.0 if ¸x) > ¸y), but 0.0 otherwise.
\pgfmathlessthan{¸x)}{¸y)}
Defines \pgfmathresult as 1.0 if ¸x) < ¸y), but 0.0 otherwise.
\pgfmathequalto{¸x)}{¸y)}
Defines \pgfmathresult 1.0 if ¸x) = ¸y), but 0.0 otherwise.
\pgfmathround{¸x)}
Defines \pgfmathresult as ¸¸x)|. This uses asymmetric half-up rounding.
\pgfmathfloor{¸x)}
Defines \pgfmathresult as ¸¸x)|.
\pgfmathceil{¸x)}
Defines \pgfmathresult as ¸¸x)|.
\pgfmathpow{¸x)}{¸y)}
Defines \pgfmathresult as ¸x)
y
. ¸y) is expected to be an integer. But it can be negative.
\pgfmathmod{¸x)}{¸y)}
Defines \pgfmathresult as ¸x) modulo ¸y).
\pgfmathmax{¸x)}{¸y)}
Defines \pgfmathresult as the maximum of ¸x) or ¸y).
\pgfmathmin{¸x)}{¸y)}
Defines \pgfmathresult as the minimum ¸x) or ¸y).
\pgfmathabs{¸x)}
Defines \pgfmathresult as absolute value of ¸x).
\pgfmathreciprocal{¸x)}
Defines \pgfmathresult as 1 ÷ ¸x).
\pgfmathexp{¸x)}
Defines \pgfmathresult as e
x
. Here, ¸x) can be a non-integer. The algorithm uses a Maclaurin series.
\pgfmathsqrt{¸x)}
Defines \pgfmathresult as

¸x).
\pgfmathveclen{¸x)}{¸y)}
Defines \pgfmathresult as

¸x)
2
+ ¸y)
2
. This uses a polynomial approximation, based on ideas due
to Rouben Rostamian.
271
40.2 Trignometric Functions
\pgfmathpi
Defines \pgfmathresult as 3.14159.
\pgfmathdeg{¸x)}
Defines \pgfmathresult as ¸x) (given in radians) converted to degrees.
\pgfmathrad{¸x)}
Defines \pgfmathresult as ¸x) (given in degrees) converted to radians.
\pgfmathsin{¸x)}
Defines \pgfmathresult as the sine of ¸x).
\pgfmathcos{¸x)}
Defines \pgfmathresult as the cosine of ¸x).
\pgfmathtan{¸x)}
Defines \pgfmathresult as the tangant of ¸x).
\pgfmathasin{¸x)}
the arcsine of ¸x).
\pgfmathacos{¸x)}
Defines \pgfmathresult as the arccosine of ¸x).
\pgfmathatan{¸x)}
Defines \pgfmathresult as the arctangent of ¸x).
40.3 Pseudo-Random Numbers
\pgfmathgeneratepseudorandomnumber
Defines \pgfmathresult as a pseudo-random integer between 1 and 2
31
−1. This uses a linear congru-
ency generator, based on ideas due to Erich Janka.
\pgfmathrnd
Defines \pgfmathresult as a pseudo-random number between 0.0 and 1.0
\pgfmathrand
Defines \pgfmathresult as a pseudo-random number between -1.0 and 1.0
\pgfmathrandominteger{¸macro)}{¸maximum)}{¸minimum)}
This defines ¸macro) as a pseudo-randomly generated integer from the range ¸maximum) to ¸minimum)
(inclusive).
\begin{pgfpicture}
\foreach \x in {1,...,50}{
\pgfmathrandominteger{\a}{1}{50}
\pgfmathrandominteger{\b}{1}{50}
\pgfpathcircle{\pgfpoint{+\a pt}{+\b pt}}{+2pt}
\color{blue!40!white}
\pgfsetstrokecolor{blue!80!black}
\pgfusepath{stroke, fill}
}
\end{pgfpicture}
\pgfmathdeclarerandomlist{¸list name)}¦{¸item-1)}{¸item 2)}...¦
This creates a list of items with the name ¸list name).
272
\pgfmathrandomitem{¸macro)}{¸list name)}
Select an item from a random list ¸list name). The selected item is placed in ¸macro).
\begin{pgfpicture}
\pgfmathdeclarerandomlist{color}{{red}{blue}{green}{yellow}{white}}
\foreach \a in {1,...,50}{
\pgfmathrandominteger{\x}{1}{85}
\pgfmathrandominteger{\y}{1}{85}
\pgfmathrandominteger{\r}{5}{10}
\pgfmathrandomitem{\c}{color}
\pgfpathcircle{\pgfpoint{+\x pt}{+\y pt}}{+\r pt}
\color{\c!40!white}
\pgfsetstrokecolor{\c!80!black}
\pgfusepath{stroke, fill}
}
\end{pgfpicture}
\pgfmathsetseed{¸integer)}
Explicitly set seed for the pseudo-random number generator. By default it is set to the value of
\time\year.
40.4 Conversion Between Bases
pgf provides limited support for conversion between representations of numbers. Currently the numbers
must be positive integers in the range 0 to 2
31
−1, and the bases in the range 2 to 36. All digits representing
numbers greater than 9 (in base ten), are alphabetic, but may be upper or lower case. Note, that again,
examples in this section, show the result of the calculation not what is displayed in the document.
\pgfmathbasetodec{¸macro)}{¸number)}{¸base)}
Defines ¸macro) as the result of converting ¸number) from base ¸base) to base 10. Alphabetic digits can
be upper or lower case.
4223 \pgfmathbasetodec\mynumber{107f}{16}
25512 \pgfmathbasetodec\mynumber{33FC}{20}
\pgfmathdectobase{¸macro)}{¸number)}{¸base)}
Defines ¸macro) as the result of converting ¸number) from base 10 to base ¸base). Any resulting
alphabetic digits are in lower case.
ffff \pgfmathdectobase\mynumber{65535}{16}
\pgfmathdectoBase{¸macro)}{¸number)}{¸base)}
Defines ¸macro) as the result of converting ¸number) from base 10 to base ¸base). Any resulting
alphabetic digits are in upper case.
FFFF \pgfmathdectoBase\mynumber{65535}{16}
\pgfmathbasetobase{¸macro)}{¸number)}{¸base-1)}{¸base-2)}
Defines ¸macro) as the result of converting ¸number) from base ¸base-1) to base ¸base-2). Alphabetic
digits in ¸number) can be upper or lower case, but any resulting alphabetic digits are in lower case.
db \pgfmathbasetobase\mynumber{11011011}{2}{16}
\pgfmathbasetoBase{¸macro)}{¸number)}{¸base-1)}{¸base-2)}
Defines ¸macro) as the result of converting ¸number) from base ¸base-1) to base ¸base-2). Alphabetic
digits in ¸number) can be upper or lower case, but any resulting alphabetic digits are in upper case.
31B \pgfmathbasetoBase\mynumber{121212}{3}{12}
273
\pgfmathsetbasenumberlength{¸integer)}
Set the number of digits in the result of a base conversion to ¸integer). If the result of a conversion has
less digits than this number it is prefixed with zeros.
00001111 \pgfmathsetbasenumberlength{8}
\pgfmathdectobase\mynumber{15}{2}
274
41 Reimplementing the Computations of the
Mathematical Engine
Perhaps you are not satisfied with the Maclaurin series for e
x
. Perhaps you have a fantastically more accurate
and efficient way of calculating the sine or cosine of angles. Perhaps you would like the library to interface
with a package such as fp for fixed-point arithmetic (but note that fp is very slow). In these case you will
want to replace the current implementations of the computations done by the mathematical engine by your
own code.
The mathematical engine was designed with such a replacement in mind. For this reason, the operations
and functions like \pgfmathadd are implemented in the following manner:
• \pgfmath¸function name)
This macro is the “public” interface for the function ¸function name). All arguments passed to this
macro are evaluated using \pgfmathparse and then passed on to the following function:
• \pgfmath¸function name)@
This macro is the “non-public” implementation of the functions algorithm (but note that, for speed,
the parser calls this macro rather than the “public” one). Arguments passed to this macro are expected
to be numbers without units. This is the macro which should be rewritten with your prize-winning
new algorithm.
The effect of \pgfmath¸function name)@ should be to set the macro \pgfmathresult to the correct value
(namely to the result of the computation without units). Furthermore, the function should have no other
side effects, that is, it should not change any global values. One way to achieve this is to use the following
code:
\def\pgfmath...@#1#2...{%
\begingroup%
... code for algorithm XXX ...
\pgfmath@returnone\pgfmath@x%
\endgroup%
}
The macro \pgfmath@returnone<macro> uses some \aftergroup magic to save result of the algorithm,
by defining \pgfmathresult as the expansion of <macro> without units. <macro> can be a macro containing
a number (with or without units), or a dimension or count (or possibly a skip) register. By performing the
algorithm within a T
E
X group, pgf registers such as \pgf@x, \pgf@y and \c@pgf@counta, \c@pgfcountb,
and so forth can be used at will. Note that current the implementation uses \pgfmath@x, \pgfmath@y, and
\c@pgfmath@counta, \c@pgfmath@countb throughout, so for consistency these should be employed. Whilst
they are currently \let to their pgf equivalents (see pgfmathutil.code.tex), this could change (as could
the pgf registers), so keeping things consistent is probably a good idea.
275
Part VII
The Basic Layer
by Till Tantau
x(t)
y(t)
−1 2
−1
1
2
3

x(t), y(t)

= (t sin
1
t
, t cos
1
t
)
(
2
π
, 0)
\begin{tikzpicture}
\draw[gray,very thin] (-1.9,-1.9) grid (2.9,3.9)
[step=0.25cm] (-1,-1) grid (1,1);
\draw[blue] (1,-2.1) -- (1,4.1); % asymptote
\draw[->] (-2,0) -- (3,0) node[right] {$x(t)$};
\draw[->] (0,-2) -- (0,4) node[above] {$y(t)$};
\foreach \pos in {-1,2}
\draw[shift={(\pos,0)}] (0pt,2pt) -- (0pt,-2pt) node[below] {$\pos$};
\foreach \pos in {-1,1,2,3}
\draw[shift={(0,\pos)}] (2pt,0pt) -- (-2pt,0pt) node[left] {$\pos$};
\fill (0,0) circle (0.064cm);
\draw[thick,parametric,domain=0.4:1.5,samples=200]
% The plot is reparameterised such that there are more samples
% near the center.
plot[id=asymptotic-example] function{(t*t*t)*sin(1/(t*t*t)),(t*t*t)*cos(1/(t*t*t))}
node[right] {$\bigl(x(t),y(t)\bigr) = (t\sin \frac{1}{t}, t\cos \frac{1}{t})$};
\fill[red] (0.63662,0) circle (2pt)
node [below right,fill=white,yshift=-4pt] {$(\frac{2}{\pi},0)$};
\end{tikzpicture}
276
42 Design Principles
This section describes the basic layer of pgf. This layer is build on top of the system layer. Whereas the
system layer just provides the absolute minimum for drawing graphics, the basic layer provides numerous
commands that make it possible to create sophisticated graphics easily and also quickly.
The basic layer does not provide a convenient syntax for describing graphics, which is left to frontends
like TikZ. For this reason, the basic layer is typically used only by “other programs.” For example, the
beamer package uses the basic layer extensively, but does not need a convenient input syntax. Rather,
speed and flexibility are needed when beamer creates graphics.
The following basic design principles underlie the basic layer:
1. Structuring into a core and several optional packages.
2. Consistently named T
E
X macros for all graphics commands.
3. Path-centered description of graphics.
4. Coordinate transformation system.
42.1 Core and Optional Packages
The basic layer consists of a core package, called pgfcore, which provides the most basic commands, and
several optional package like pgfbaseshade that offer more special-purpose commands.
\usepackage{pgfbaseplot} % L
A
T
E
X
\input pgfbaseplot.tex % plain T
E
X
\usemodule[pgfbaseplot] % ConT
E
Xt
provides commands for plotting functions
\usepackage{pgfbaseshapes} % L
A
T
E
X
\input pgfbaseshapes.tex % plain T
E
X
\usemodule[pgfbaseshapes] % ConT
E
Xt
provides commands for drawing shapes and nodes
\usepackage{pgfbasepatterns} % L
A
T
E
X
\input pgfbasepatterns.tex % plain T
E
X
\usemodule[pgfbasepatterns] % ConT
E
Xt
provides commands for declaring and using tiling patterns
\usepackage{pgfbaseimage} % L
A
T
E
X
\input pgfbaseimage.tex % plain T
E
X
\usemodule[pgfbaseimage] % ConT
E
Xt
This package provides commands for including external images. For L
A
T
E
X-users the graphicx pack-
age does a better job at this than the pgfbaseimage package does, so you should normally use
\includegraphics and not \pgfimage. However, in some situations (like when masking is needed
or when plain T
E
X is used) this package is needed.
\usepackage{pgfbaselayers} % L
A
T
E
X
\input pgfbaselayers.tex % plain T
E
X
\usemodule[pgfbaselayers] % ConT
E
Xt
This package provides commands for creating layered graphics. Using layers you can later on say that
a certain path should be behind a path that was specified earlier.
\usepackage{pgfbasesnakes} % L
A
T
E
X
\input pgfbasesnakes.tex % plain T
E
X
\usemodule[pgfbasesnakes] % ConT
E
Xt
This package provides commands for adding snaked lines to the path. Such lines are not straight but
rather wind in some specific fashion.
\usepackage{pgfbasematrix} % L
A
T
E
X
277
\input pgfbasematrix.tex % plain T
E
X
\usemodule[pgfbasematrix] % ConT
E
Xt
This package provides the \pgfmatrix command.
If you say \usepackage{pgf} or \input pgf.tex or \usemodule[pgf], all of the optional packages are
loaded (as well as the core and the system layer).
42.2 Communicating with the Basic Layer via Macros
In order to “communicate” with the basic layer you use long sequences of commands that start with
\pgf. You are only allowed to give these commands inside a {pgfpicture} environment. (Note that
{tikzpicture} opens a {pgfpicture} internally, so you can freely mix pgf commands and TikZ com-
mands inside a {tikzpicture}.) It is possible to “do other things” between the commands. For example,
you might use one command to move to a certain point, then have a complicated computation of the next
point, and then move there.
\newdimen\myypos
\begin{pgfpicture}
\pgfpathmoveto{\pgfpoint{0cm}{\myypos}}
\pgfpathlineto{\pgfpoint{1cm}{\myypos}}
\advance \myypos by 1cm
\pgfpathlineto{\pgfpoint{1cm}{\myypos}}
\pgfpathclose
\pgfusepath{stroke}
\end{pgfpicture}
The following naming conventions are used in the basic layer:
1. All commands and environments start with pgf.
2. All commands that specify a point (a coordinate) start with \pgfpoint.
3. All commands that extend the current path start with \pgfpath.
4. All commands that set/change a graphics parameter start with \pgfset.
5. All commands that use a previously declared object (like a path, image or shading) start with \pgfuse.
6. All commands having to do with coordinate transformations start with \pgftransform.
7. All commands having to do with arrow tips start with \pgfarrows.
8. All commands for “quickly” extending or drawing a path start with \pgfpathq or \pgfusepathq.
9. All commands having to do with matrices start with \pgfmatrix.
42.3 Path-Centered Approach
In pgf the most important entity is the path. All graphics are composed of numerous paths that can be
stroked, filled, shaded, or clipped against. Paths can be closed or open, they can self-intersect and consist
of unconnected parts.
Paths are first constructed and then used. In order to construct a path, you can use commands starting
with \pgfpath. Each time such a command is called, the current path is extended in some way.
Once a path has been completely constructed, you can use it using the command \pgfusepath. De-
pending on the parameters given to this command, the path will be stroked (drawn) or filled or subsequent
drawings will be clipped against this path.
42.4 Coordinate Versus Canvas Transformations
pgf provides two transformation systems: pgf’s own coordinate transformation matrix and pdf’s or
PostScript’s canvas transformation matrix. These two systems are quite different. Whereas a scaling by a
factor of, say, 2 of the canvas causes everything to be scaled by this factor (including the thickness of lines
and text), a scaling of two in the coordinate system causes only the coordinates to be scaled, but not the
line width nor text.
278
By default, all transformations only apply to the coordinate transformation system. However, using the
command \pgflowlevel it is possible to apply a transformation to the canvas.
Coordinate transformations are often preferable over canvas transformations. Text and lines that are
transformed using canvas transformations suffer from differing sizes and lines whose thickness differs de-
pending on whether the line is horizontal or vertical. To appreciate the difference, consider the following
two “circles” both of which have been scaled in the x-direction by a factor of 3 and by a factor of 0.5 in
the y-direction. The left circle uses a canvas transformation, the right uses pgf’s coordinate transformation
(some viewers will render the left graphic incorrectly since they do no apply the low-level transformation the
way they should):
canvas
coordinate
279
43 Hierarchical Structures:
Package, Environments, Scopes, and Text
43.1 Overview
pgf uses two kinds of hierarchical structuring: First, the package itself is structured hierarchically, consisting
of different packages that are built on top of each other. Second, pgf allows you to structure your graphics
hierarchically using environments and scopes.
43.1.1 The Hierarchical Structure of the Package
The pgf system consists of several layers:
System layer. The lowest layer is called the system layer, though it might also be called “driver layer”
or perhaps “backend layer.” Its job is to provide an abstraction of the details of which driver is used
to transform the .dvi file. The system layer is implemented by the package pgfsys, which will load
appropriate driver files as needed.
The system layer is documented in Part VIII.
Basic layer. The basic layer is loaded by the package pgf. Some applications do not need all of the
functionality of the basic layer, so it is possible to load only the pgfcore and some other packages
starting with pgfbase.
The basic layer is documented in the present part.
Frontend layer. The frontend layer is not loaded by a single packages. Rather, different packages, like
TikZ or pgfpict2e, are different frontends to the basic layer.
The TikZ frontend is documented in Part III.
Each layer will automatically load the necessary files of the layers below it.
In addition to the packages of these layers, there are also some library packages. These packages provide
additional definitions of things like new arrow tips or new plot handlers.
The library packages are documented in Part IV.
43.1.2 The Hierarchical Structure of Graphics
Graphics in pgf are typically structured hierarchically. Hierarchical structuring can be used to identify
groups of graphical elements that are to be treated “in the same way.” For example, you might group
together a number of paths, all of which are to be drawn in red. Then, when you decide later on that you
like them to be drawn in, say, blue, all you have to do is to change the color once.
The general mechanism underlying hierarchical structuring is known as scoping in computer science.
The idea is that all changes to the general “state” of the graphic that are done inside a scope are local to
that scope. So, if you change the color inside a scope, this does not affect the color used outside the scope.
Likewise, when you change the line width in a scope, the line width outside is not changed, and so on.
There are different ways of starting and ending scopes of graphic parameters. Unfortunately, these scopes
are sometimes “in conflict” with each other and it is sometimes not immediately clear which scopes apply.
In essence, the following scoping mechanisms are available:
1. The “outermost” scope supported by pgf is the {pgfpicture} environment. All changes to the graphic
state done inside a {pgfpicture} are local to that picture.
In general, it is not possible to set graphic parameters globally outside any {pgfpicture} environments.
Thus, you can not say \pgfsetlinewidth{1pt} at the beginning of your document to have a default
line width of one point. Rather, you have to (re)set all graphic parameters inside each {pgfpicture}.
(If this is too bothersome, try defining some macro that does the job for you.)
2. Inside a {pgfpicture} you can use a {pgfscope} environment to keep changes of the graphic state
local to that environment.
The effect of commands that change the graphic state are local to the current {pgfscope} but not
always to the current T
E
X group. Thus, if you open a T
E
X group (some text in curly braces) inside a
280
{pgfscope}, and if you change, for example, the dash pattern, the effect of this changed dash pattern
will persist till the end of the {pgfscope}.
Unfortunately, this is not always the case. Some graphic parameters only persist till the end of the
current T
E
X group. For example, when you use \pgfsetarrows to set the arrow tip inside a T
E
X
group, the effect lasts only till the end of the current T
E
X group.
3. Some graphic parameters are not scoped by {pgfscope} but “already” by T
E
X groups. For example,
the effect of coordinate transformation commands is always local to the current T
E
X group.
Since every {pgfscope} automatically creates a T
E
X group, all graphic parameters that are local to
the current T
E
X group are also local to the current {pgfscope}.
4. Some graphic parameters can only be scoped using T
E
X groups, since in some situations it is not
possible to introduce a {pgfscope}. For example, a path always has to be completely constructed and
used in the same {pgfscope}. However, we might wish to have different coordinate transformations
apply to different points on the path. In this case, we can use T
E
X groups to keep the effect local, but
we could not use {pgfscope}.
5. The \pgftext command can be used to create a scope in which T
E
X “escapes back” to normal T
E
X
mode. The text passed to the \pgftext is “heavily guarded” against having any effect on the scope
in which it is used. For example, it is possibly to use another {pgfpicture} environment inside the
argument of \pgftext.
Most of the complications can be avoided if you stick to the following rules:
• Give graphic commands only inside {pgfpicture} environments.
• Use {pgfscope} to structure graphics.
• Do not use T
E
X groups inside graphics, except for keeping the effect of coordinate transformations
local.
43.2 The Hierarchical Structure of the Package
Before we come to the structuring commands provided by pgf to structure your graphics, let us first have
a look at the structure of the package itself.
43.2.1 The Main Package
To use pgf, include the following package:
\usepackage{pgf} % L
A
T
E
X
\input pgf.tex % plain T
E
X
\usemodule[pgf] % ConT
E
Xt
This package loads the complete “basic layer” of pgf. That is, it will load all of the commands described
in the current part of this manual, but it will not load frontends like TikZ.
In detail, this package will load the following packages, each of which can also be loaded individually:
• pgfsys, which is the lowest layer of pgf and which is always needed. This file will read pgf.cfg
to discern which driver is to be used. See Section 59.1 for details.
• pgfcore, which is the central core of pgf and which is always needed unless you intend to write
a new basic layer from scratch.
• pgfbaseimage, which provides commands for declaring and using images. An example is
\pgfuseimage.
• pgfbaseshapes, which provides commands for declaring and using shapes. An example is
\pgfdeclareshape.
• pgfbaseplot, which provides commands for plotting functions.
Including any of the last three packages will automatically load the first two.
In L
A
T
E
X, the package takes two options:
281
\usepackage[draft]{pgf}
When this option is set, all images will be replaced by empty rectangles. This can speedup compilation.
\usepackage[version=¸version )]{pgf}
Indicates that the commands of version ¸version) need to be defined. If you set ¸version) to 0.65,
then a large bunch of “compatibility commands” are loaded. If you set ¸version) to 0.96, then these
compatibility commands will not be loaded.
If this option is not given at all, then the commands of all versions are defined.
43.2.2 The Core Package
\usepackage{pgfcore} % L
A
T
E
X
\input pgfcore.tex % plain T
E
X
\usemodule[pgfcore] % ConT
E
Xt
This package defines all of the basic layer’s commands, except for the commands defined in the addi-
tional packages like pgfbaseplot. Typically commands defined by the core include \pgfusepath or
\pgfpoint. The core is internally structured into several subpackages, but the subpackages cannot be
loaded individually since they are all “interrelated.”
43.2.3 The Optional Basic Layer Packages
The pgf package automatically loads the following packages, but you can also load them individually (all of
them automatically include the core):
• pgfbaseshapes This package provides commands for drawing nodes and shapes. These commands are
explained in Section 49.
• pgfbaseplot This package provides commands for plotting function. The commands are explained in
Section 56.
• pgfbaseimage This package provides commands for including (external) images. The commands are
explained in Section 53.
43.2.4 The Library Packages
There is a special command for loading library packages.
\usepgflibrary{¸list of libraries)}
Use this command to load further libraries. The list of libraries should contain the names of libraries
separated by commas. Instead of curly braces, you can also use square brackets. If you try to load a
library a second time, nothing will happen.
Example: \usepgflibrary{arrows}
This command causes the the file pgflibrary¸library).code.tex to be loaded for each ¸library) in the
¸list of libraries). This means that in order to write your own library file, place a file of the appropriate
name somewhere where T
E
X can find it. L
A
T
E
X, plain T
E
X, and ConT
E
Xt users can then use your
library.
You should also consider adding a TikZ library that simply includes your pgf library.
43.3 The Hierarchical Structure of the Graphics
43.3.1 The Main Environment
Most, but not all, commands of the pgf package must be given within a {pgfpicture} environment.
The only commands that (must) be given outside are commands having to do with including images (like
\pgfuseimage) and with inserting complete shadings (like \pgfuseshading). However, just to keep life
entertaining, the \pgfshadepath command must be given inside a {pgfpicture} environment.
\begin{pgfpicture}
¸environment contents)
282
\end{pgfpicture}
This environment will insert a T
E
X box containing the graphic drawn by the ¸environment contents) at
the current position.
The size of the bounding box. The size of the box is determined in the following manner: While
pgf parses the ¸environment contents), it keeps track of a bounding box for the graphic. Essentially,
this bounding box is the smallest box that contains all coordinates mentioned in the graphics. Some
coordinates may be “mentioned” by pgf itself; for example, when you add circle to the current path,
the support points of the curve making up the circle are also “mentioned” despite the fact that you will
not “see” them in your code.
Once the ¸environment contents) has been parsed completely, a T
E
X box is created whose size is the
size of the computed bounding box and this box is inserted at the current position.
Hello World!
Hello \begin{pgfpicture}
\pgfpathrectangle{\pgfpointorigin}{\pgfpoint{2ex}{1ex}}
\pgfusepath{stroke}
\end{pgfpicture} World!
Sometimes, you may need more fine-grained control over the size of the bounding box. For example, the
computed bounding box may be too large or you intensionally wish the box to be “too small.” In these
cases, you can use the command \pgfusepath{use as bounding box}, as described in Section 47.5.
The baseline of the bounding box. When the box containing the graphic is inserted into the
normal text, the baseline of the graphic is normally at the bottom of the graphic. For this reason,
the following two sets of code lines have the same effect, despite the fact that the second graphic uses
“higher” coordinates than the first:
Rectangles and .
Rectangles \begin{pgfpicture}
\pgfpathrectangle{\pgfpointorigin}{\pgfpoint{2ex}{1ex}}
\pgfusepath{stroke}
\end{pgfpicture} and \begin{pgfpicture}
\pgfpathrectangle{\pgfpoint{0ex}{1ex}}{\pgfpoint{2ex}{1ex}}
\pgfusepath{stroke}
\end{pgfpicture}.
You can change the baseline using the \pgfsetbaseline command, see below.
Rectangles and .
Rectangles \begin{pgfpicture}
\pgfpathrectangle{\pgfpointorigin}{\pgfpoint{2ex}{1ex}}
\pgfusepath{stroke}
\pgfsetbaseline{0pt}
\end{pgfpicture} and \begin{pgfpicture}
\pgfpathrectangle{\pgfpoint{0ex}{1ex}}{\pgfpoint{2ex}{1ex}}
\pgfusepath{stroke}
\pgfsetbaseline{0pt}
\end{pgfpicture}.
Including text and images in a picture. You cannot directly include text and images in a
picture. Thus, you should not simply write some text in a {pgfpicture} or use a command like
\includegraphics or even \pgfimage. In all these cases, you need to place the text inside a \pgftext
command. This will “escape back” to normal T
E
X mode, see Section 43.3.3 for details.
Remembering a picture position for later reference. After a picture has been typset, its position
on the page is normally forgotten by pgf and also by T
E
X. This means that is not possible to reference
a node in this picture later on. In particular, it is normally impossible to draw lines between nodes in
different pictures automatically.
In order to make pgf “remember” a picture, the T
E
X-if \ifpgfrememberpicturepositiononpage
should be set to true. It is only important that this T
E
X-if is true at the end of the {pgfpicture}-
environment, so you can switch it on inside the environment. However, you can also just switch it on
globally, then the positions of all pictures are remembered.
There are several reasons why the remembering is not switched on by default. First, it does not work
for all backend drivers (currently, it works only for pdfT
E
X). Second, it requires two passes of T
E
X over
283
the file; on the first pass all positions will be wrong. Third, for every remembered picture a line is added
to the .aux-file, which may result in a large number of extra lines.
Despite all these “problems,” for documents that are processed with pdfT
E
X and in which there is only
a small number of pictures (less than a hundred or so), you can switch on this option globally, it will
not cause any significant slowing of T
E
X.
\pgfpicture
¸environment contents)
\endpgfpicture
The plain T
E
X version of the environment. Note that in this version, also, a T
E
X group is created
around the environment.
\startpgfpicture
¸environment contents)
\stoppgfpicture
This is the ConT
E
Xt version of the environment.
\ifpgfrememberpicturepositiononpage
Determines whether the position of pictures on the page should be recorded. The value of this T
E
X-if
at the end of a {pgfpicture} environment is important, not the value at the beginning.
If this option is set to true of a picture, pgf will attempt to record the position of the picture on the
page. (This attempt will fail with most drivers and when it works it typically requires two runs of T
E
X.)
The position is not directly accessible. Rather, the nodes mechanism will use this position if you access
a node from another picture. See Sections 49.3.2 and 13.12 for more details.
\pgfsetbaseline{¸dimension)}
This command specifies a y-coordinate of the picture that should be used as the baseline of the whole
picture. When a pgf picture has been typeset completely, pgf must decide at which height the baseline
of the picture should lie. Normally, the baseline is set to the y-coordinate of the bottom of the picture,
but it is often desirable to use another height.
Text , , , .
Text \tikz{\pgfpathcircle{\pgfpointorigin}{1ex}\pgfusepath{stroke}},
\tikz{\pgfsetbaseline{0pt}
\pgfpathcircle{\pgfpointorigin}{1ex}\pgfusepath{stroke}},
\tikz{\pgfsetbaseline{.5ex}
\pgfpathcircle{\pgfpointorigin}{1ex}\pgfusepath{stroke}},
\tikz{\pgfsetbaseline{-1ex}
\pgfpathcircle{\pgfpointorigin}{1ex}\pgfusepath{stroke}}.
\pgfsetbaselinepointnow{¸point)}
This command specifies the baseline indirectly, namely as the y-coordinate that the given ¸point) has
when the command is called.
\pgfsetbaselinepointlater{¸point)}
This command also specifies the baseline indirectly, but the y-coordinate of the given ¸point) is only
computed at the end of the picture.
Hello world.
Hello
\tikz{
\pgfsetbaselinepointlater{\pgfpointanchor{X}{base}}
% Note: no shape X, yet
\node [cross out,draw] (X) {world.};
}
43.3.2 Graphic Scope Environments
Inside a {pgfpicture} environment you can substructure your picture using the following environment:
\begin{pgfscope}
¸environment contents)
284
\end{pgfscope}
All changes to the graphic state done inside this environment are local to the environment. The graphic
state includes the following:
• The line width.
• The stroke and fill colors.
• The dash pattern.
• The line join and cap.
• The miter limit.
• The canvas transformation matrix.
• The clipping path.
Other parameters may also influence how graphics are rendered, but they are not part of the graphic
state. For example, the arrow tip kind is not part of the graphic state and the effect of commands
setting the arrow tip kind are local to the current T
E
X group, not to the current {pgfscope}. However,
since {pgfscope} starts and ends a T
E
X group automatically, a {pgfscope} can be used to limit the
effect of, say, commands that set the arrow tip kind.
\begin{pgfpicture}
\begin{pgfscope}
{
\pgfsetlinewidth{2pt}
\pgfpathrectangle{\pgfpointorigin}{\pgfpoint{2ex}{2ex}}
\pgfusepath{stroke}
}
\pgfpathrectangle{\pgfpoint{3ex}{0ex}}{\pgfpoint{2ex}{2ex}}
\pgfusepath{stroke}
\end{pgfscope}
\pgfpathrectangle{\pgfpoint{6ex}{0ex}}{\pgfpoint{2ex}{2ex}}
\pgfusepath{stroke}
\end{pgfpicture}
\begin{pgfpicture}
\begin{pgfscope}
{
\pgfsetarrows{-to}
\pgfpathmoveto{\pgfpointorigin}\pgfpathlineto{\pgfpoint{2ex}{2ex}}
\pgfusepath{stroke}
}
\pgfpathmoveto{\pgfpoint{3ex}{0ex}}\pgfpathlineto{\pgfpoint{5ex}{2ex}}
\pgfusepath{stroke}
\end{pgfscope}
\pgfpathmoveto{\pgfpoint{6ex}{0ex}}\pgfpathlineto{\pgfpoint{8ex}{2ex}}
\pgfusepath{stroke}
\end{pgfpicture}
At the start of the scope, the current path must be empty, that is, you cannot open a scope while
constructing a path.
It is usually a good idea not to introduce T
E
X groups inside a {pgfscope} environment.
\pgfscope
¸environment contents)
\endpgfscope
Plain T
E
X version of the {pgfscope} environment.
\startpgfscope
¸environment contents)
\stoppgfscope
This is the ConT
E
Xt version of the environment.
The following scopes also encapsulate certain properties of the graphic state. However, they are typically
not used directly by the user.
285
\begin{pgfinterruptpath}
¸environment contents)
\end{pgfinterruptpath}
This environment can be used to temporarily interrupt the construction of the current path. The effect
will be that the path currently under construction will be “stored away” and restored at the end of the
environment. Inside the environment you can construct a new path and do something with it.
An example application of this environment is the arrow tip caching. Suppose you ask pgf to use a
specific arrow tip kind. When the arrow tip needs to be rendered for the first time, pgf will “cache”
the path that makes up the arrow tip. To do so, it interrupts the current path construction and then
protocols the path of the arrow tip. The {pgfinterruptpath} environment is used to ensure that this
does not interfere with the path to which the arrow tips should be attached.
This command does not install a {pgfscope}. In particular, it does not call any \pgfsys@ commands
at all, which would, indeed, be dangerous in the middle of a path construction.
\pgfinterruptpath
¸environment contents)
\endpgfinterruptpath
Plain T
E
X version of the environment.
\startpgfinterruptpath
¸environment contents)
\stoppgfinterruptpath
ConT
E
Xt version of the environment.
\begin{pgfinterruptpicture}
¸environment contents)
\end{pgfinterruptpicture}
This environment can be used to temporarily interrupt a {pgfpicture}. However, the environment is
intended only to be used at the beginning and end of a box that is (later) inserted into a {pgfpicture}
using \pgfqbox. You cannot use this environment directly inside a {pgfpicture}.
Sub--picture.
\begin{pgfpicture}
\pgfpathmoveto{\pgfpoint{0cm}{0cm}} % In the middle of path, now
\newbox\mybox
\setbox\mybox=\hbox{
\begin{pgfinterruptpicture}
Sub-\begin{pgfpicture} % a subpicture
\pgfpathmoveto{\pgfpoint{1cm}{0cm}}
\pgfpathlineto{\pgfpoint{1cm}{1cm}}
\pgfusepath{stroke}
\end{pgfpicture}-picture.
\end{pgfinterruptpicture}
}
\pgfqbox{\mybox}%
\pgfpathlineto{\pgfpoint{0cm}{1cm}}
\pgfusepath{stroke}
\end{pgfpicture}\hskip3.9cm
\pgfinterruptpicture
¸environment contents)
\endpgfinterruptpicture
Plain T
E
X version of the environment.
\startpgfinterruptpicture
¸environment contents)
\stoppgfinterruptpicture
ConT
E
Xt version of the environment.
\begin{pgfinterruptboundingbox}
¸environment contents)
286
\end{pgfinterruptboundingbox}
This environment temporarily interrupts the computation of the bounding box and sets up a new
bounding box. At the beginning of the environment the old bounding box is saved and an empty
bounding box is installed. After the environment the orginal bounding box is reinstalled as if nothing
has happened.
\pgfinterruptboundingbox
¸environment contents)
\endpgfinterruptboundingbox
Plain T
E
X version of the environment.
\startpgfinterruptboundingbox
¸environment contents)
\stoppgfinterruptboundingbox
ConT
E
Xt version of the environment.
43.3.3 Inserting Text and Images
Often, you may wish to add normal T
E
X text at a certain point inside a {pgfpicture}. You cannot do so
“directly,” that is, you cannot simply write this text inside the {pgfpicture} environment. Rather, you
must pass the text as an argument to the \pgftext command.
You must also use the \pgftext command to insert an image or a shading into a {pgfpicture}.
\pgftext[¸options)]{¸text)}
This command will typeset ¸text) in normal T
E
X mode and insert the resulting box into the
{pgfpicture}. The bounding box of the graphic will be updated so that all of the text box is in-
side. Be default, the text box is centered at the origin, but this can be changed either by giving
appropriate ¸options) or by applying an appropriate coordinate transformation beforehand.
The ¸text) may contain verbatim text. (In other words, the ¸text) “argument” is not a normal argument,
but is put in a box and some \aftergroup hackery is used to find the end of the box.)
pgf’s current (high-level) coordinate transformation is synchronized with the canvas transformation
matrix temporarily when the text box is inserted. The effect is that if there is currently a high-level
rotation of, say, 30 degrees, the ¸text) will also be rotated by thirty degrees. If you do not want this
effect, you have to (possibly temporarily) reset the high-level transformation matrix.
The following ¸options) may be given as conveniences:
• left causes the text box to be placed such that its left border is on the origin.
lovely
\tikz{\draw[help lines] (-1,-.5) grid (1,.5);
\pgftext[left] {lovely}}
• right causes the text box to be placed such that its right border is on the origin.
lovely
\tikz{\draw[help lines] (-1,-.5) grid (1,.5);
\pgftext[right] {lovely}}
• top causes the text box to be placed such that its top is on the origin. This option can be used
together with the left or right option.
lovely
\tikz{\draw[help lines] (-1,-.5) grid (1,.5);
\pgftext[top] {lovely}}
lovely
\tikz{\draw[help lines] (-1,-.5) grid (1,.5);
\pgftext[top,right] {lovely}}
287
• bottom causes the text box to be placed such that its bottom is on the origin.
lovely
\tikz{\draw[help lines] (-1,-.5) grid (1,.5);
\pgftext[bottom] {lovely}}
lovely
\tikz{\draw[help lines] (-1,-.5) grid (1,.5);
\pgftext[bottom,right] {lovely}}
• base causes the text box to be placed such that its baseline is on the origin.
lovely
\tikz{\draw[help lines] (-1,-.5) grid (1,.5);
\pgftext[base] {lovely}}
lovely
\tikz{\draw[help lines] (-1,-.5) grid (1,.5);
\pgftext[base,right] {lovely}}
• at=¸point) Translates the origin (that is, the point where the text is shown) to ¸point).
lovely
\tikz{\draw[help lines] (-1,-.5) grid (1,.5);
\pgftext[base,at={\pgfpoint{1cm}{0cm}}] {lovely}}
• x=¸dimension) Translates the origin by ¸dimension) along the x-axis.
lovely
\tikz{\draw[help lines] (-1,-.5) grid (1,.5);
\pgftext[base,x=1cm,y=-0.5cm] {lovely}}
• y=¸dimension) works like the x option.
• rotate=¸degree) Rotates the coordinate system by ¸degree). This will also rotate the text box.
l
o
v
e
l
y
\tikz{\draw[help lines] (-1,-.5) grid (1,.5);
\pgftext[base,x=1cm,y=-0.5cm,rotate=30] {lovely}}
288
44 Specifying Coordinates
44.1 Overview
Most pgf commands expect you to provide the coordinates of a point (also called coordinate) inside your
picture. Points are always “local” to your picture, that is, they never refer to an absolute position on the
page, but to a position inside the current {pgfpicture} environment. To specify a coordinate you can use
commands that start with \pgfpoint.
44.2 Basic Coordinate Commands
The following commands are the most basic for specifying a coordinate.
\pgfpoint{¸x coordinate)}{¸y coordinate)}
Yields a point location. The coordinates are given as T
E
X dimensions.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathcircle{\pgfpoint{1cm}{1cm}} {2pt}
\pgfpathcircle{\pgfpoint{2cm}{5pt}} {2pt}
\pgfpathcircle{\pgfpoint{0pt}{.5in}}{2pt}
\pgfusepath{fill}
\end{tikzpicture}
\pgfpointorigin
Yields the origin. Same as \pgfpoint{0pt}{0pt}.
\pgfpointpolar{¸degree)}{¸radius)/¸y-radius)}
Yields a point location given in polar coordinates. You can specify the angle only in degrees, radians
are not supported, currently.
If the optional ¸y-radius) is given, the polar coordinate is actually a coordinate on an ellipse whose
x-radius is given by ¸radius) and whose y-radius is given by ¸y-radius).
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\foreach \angle in {0,10,...,90}
{\pgfpathcircle{\pgfpointpolar{\angle}{1cm}}{2pt}}
\pgfusepath{fill}
\end{tikzpicture}
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\foreach \angle in {0,10,...,90}
{\pgfpathcircle{\pgfpointpolar{\angle}{1cm/2cm}}{2pt}}
\pgfusepath{fill}
\end{tikzpicture}
44.3 Coordinates in the XY-Coordinate System
Coordinates can also be specified as multiples of an x-vector and a y-vector. Normally, the x-vector points
one centimeter in the x-direction and the y-vector points one centimeter in the y-direction, but using the
commands \pgfsetxvec and \pgfsetyvec they can be changed. Note that the x- and y-vector do not
necessarily point “horizontally” and “vertically.”
\pgfpointxy{¸s
x
)}{¸s
y
)}
Yields a point that is situated at s
x
times the x-vector plus s
y
times the y-vector.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathmoveto{\pgfpointxy{1}{0}}
\pgfpathlineto{\pgfpointxy{2}{2}}
\pgfusepath{stroke}
\end{tikzpicture}
289
\pgfsetxvec{¸point)}
Sets that current x-vector for usage in the xyz-coordinate system.
Example:
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathmoveto{\pgfpointxy{1}{0}}
\pgfpathlineto{\pgfpointxy{2}{2}}
\pgfusepath{stroke}
\color{red}
\pgfsetxvec{\pgfpoint{0.75cm}{0cm}}
\pgfpathmoveto{\pgfpointxy{1}{0}}
\pgfpathlineto{\pgfpointxy{2}{2}}
\pgfusepath{stroke}
\end{tikzpicture}
\pgfsetyvec{¸point)}
Works like \pgfsetyvec.
\pgfpointpolarxy{¸degree)}{¸radius)/¸y-radius)}
This command is similar to the \pgfpointpolar command, but the ¸radius) is now a factor to be
interpreted in the xy-coordinate system. This means that a degree of 0 is the same as the x-vector
of the xy-coordinate system times ¸radius) and a degree of 90 is the y-vecotr times ¸radius). As for
\pgfpointpolar, a ¸radius) can also be a pair separated by a slash. In this case, the x- and y-vectors
are multiplied by different factors.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\begin{scope}[x={(1cm,-5mm)},y=1.5cm]
\foreach \angle in {0,10,...,90}
{\pgfpathcircle{\pgfpointpolarxy{\angle}{1}}{2pt}}
\pgfusepath{fill}
\end{scope}
\end{tikzpicture}
44.4 Three Dimensional Coordinates
It is also possible to specify a point as a multiple of three vectors, the x-, y-, and z-vector. This is useful for
creating simple three dimensional graphics.
\pgfpointxyz{¸s
x
)}{¸s
y
)}{¸s
z
)}
Yields a point that is situated at s
x
times the x-vector plus s
y
times the y-vector plus s
z
times the
z-vector.
\begin{pgfpicture}
\pgfsetarrowsend{to}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpointxyz{0}{0}{1}}
\pgfusepath{stroke}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpointxyz{0}{1}{0}}
\pgfusepath{stroke}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpointxyz{1}{0}{0}}
\pgfusepath{stroke}
\end{pgfpicture}
\pgfsetzvec{¸point)}
Works like \pgfsetzvec.
290
Inside the xyz-coordinate system, you can also specify points using spherical and cylindrical coordinates.
\pgfpointcylindrical{¸degree)}{¸radius)}{¸height)}
This command yields the same as
\pgfpointadd{\pgfpointpolarxy{degree}{radius}}{\pgfpointxyz{0}{0}{height}}
x
y
z
\begin{tikzpicture}
\draw [->] (0,0) -- (1,0,0) node [right] {$x$};
\draw [->] (0,0) -- (0,1,0) node [above] {$y$};
\draw [->] (0,0) -- (0,0,1) node [below left] {$z$};
\pgfpathcircle{\pgfpointcylindrical{80}{1}{.5}}{2pt}
\pgfusepath{fill}
\draw[red] (0,0) -- (0,0,.5) -- +(80:1);
\end{tikzpicture}
\pgfpointspherical{¸longitude)}{¸latitude)}{¸radius)}
This command yields a point “on the surface of the earth” specified by the ¸longitude) and the
{¸latitude)}. The radius of the earth is given by ¸radius). The equator lies in the xy-plane.
\begin{tikzpicture}
\pgfsetfillcolor{lightgray}
\foreach \latitude in {-90,-75,...,30}
{
\foreach \longitude in {0,20,...,360}
{
\pgfpathmoveto{\pgfpointspherical{\longitude}{\latitude}{1}}
\pgfpathlineto{\pgfpointspherical{\longitude+20}{\latitude}{1}}
\pgfpathlineto{\pgfpointspherical{\longitude+20}{\latitude+15}{1}}
\pgfpathlineto{\pgfpointspherical{\longitude}{\latitude+15}{1}}
\pgfpathclose
}
\pgfusepath{fill,stroke}
}
\end{tikzpicture}
44.5 Building Coordinates From Other Coordinates
Many commands allow you to construct a coordinate in terms of other coordinates.
44.5.1 Basic Manipulations of Coordinates
\pgfpointadd{¸v
1
)}{¸v
2
)}
Returns the sum vector ¸v
1
) + ¸v
2
).
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathcircle{\pgfpointadd{\pgfpoint{1cm}{0cm}}{\pgfpoint{1cm}{1cm}}}{2pt}
\pgfusepath{fill}
\end{tikzpicture}
\pgfpointscale{¸factor)}{¸coordinate)}
Returns the vector ¸factor)¸coordinate).
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathcircle{\pgfpointscale{1.5}{\pgfpoint{1cm}{0cm}}}{2pt}
\pgfusepath{fill}
\end{tikzpicture}
291
\pgfpointdiff{¸start)}{¸end)}
Returns the difference vector ¸end) − ¸start).
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathcircle{\pgfpointdiff{\pgfpoint{1cm}{0cm}}{\pgfpoint{1cm}{1cm}}}{2pt}
\pgfusepath{fill}
\end{tikzpicture}
\pgfpointnormalised{¸point)}
This command returns a normalized version of ¸point), that is, a vector of length 1pt pointing in the
direction of ¸point). If ¸point) is the 0-vector or extremely short, a vector of length 1pt pointing upwards
is returned.
This command is not implemented by calculating the length of the vector, but rather by calculating
the angle of the vector and then using (something equivalent to) the \pgfpointpolar command. This
ensures that the point will really have length 1pt, but it is not guaranteed that the vector will precisely
point in the direction of ¸point) due to the fact that the polar tables are accurate only up to one degree.
Normally, this is not a problem.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathcircle{\pgfpoint{2cm}{1cm}}{2pt}
\pgfpathcircle{\pgfpointscale{20}
{\pgfpointnormalised{\pgfpoint{2cm}{1cm}}}}{2pt}
\pgfusepath{fill}
\end{tikzpicture}
44.5.2 Points Traveling along Lines and Curves
The commands in this section allow you to specify points on a line or a curve. Imaging a point “traveling”
along a curve from some point p to another point q. At time t = 0 the point is at p and at time t = 1 it
is at q and at time, say, t = 1/2 it is “somewhere in the middle.” The exact location at time t = 1/2 will
not necessarily be the “halfway point,” that is, the point whose distance on the curve from p and q is equal.
Rather, the exact location will depend on the “speed” at which the point is traveling, which in turn depends
on the lengths of the support vectors in a complicated manner. If you are interested in the details, please
see a good book on B´ezier curves.
\pgfpointlineattime{¸time t)}{¸point p)}{¸point q)}
Yields a point that is the tth fraction between p and q, that is, p + t(q − p). For t = 1/2 this is the
middle of p and q.
0
0.25
0.5
0.75
1
1.25 \begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{2cm}{2cm}}
\pgfusepath{stroke}
\foreach \t in {0,0.25,...,1.25}
{\pgftext[at=
\pgfpointlineattime{\t}{\pgfpointorigin}{\pgfpoint{2cm}{2cm}}]{\t}}
\end{tikzpicture}
\pgfpointlineatdistance{¸distance)}{¸start point)}{¸end point)}
Yields a point that is located ¸distance) many units removed from the start point in the direction of the
end point. In other words, this is the point that results if we travel ¸distance) steps from ¸start point)
towards ¸end point).
Example:
292
0pt
20pt
40pt
70pt
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{3cm}{2cm}}
\pgfusepath{stroke}
\foreach \d in {0pt,20pt,40pt,70pt}
{\pgftext[at=
\pgfpointlineatdistance{\d}{\pgfpointorigin}{\pgfpoint{3cm}{2cm}}]{\d}}
\end{tikzpicture}
\pgfpointcurveattime{¸time t)}{¸point p)}{¸point s
1
)}{¸point s
2
)}{¸point q)}
Yields a point that is on the B´ezier curve from p to q with the support points s
1
and s
2
. The time t is
used to determine the location, where t = 0 yields p and t = 1 yields q.
0
0.25
0.5
0.75 1 \begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathmoveto{\pgfpointorigin}
\pgfpathcurveto
{\pgfpoint{0cm}{2cm}}{\pgfpoint{0cm}{2cm}}{\pgfpoint{3cm}{2cm}}
\pgfusepath{stroke}
\foreach \t in {0,0.25,0.5,0.75,1}
{\pgftext[at=\pgfpointcurveattime{\t}{\pgfpointorigin}
{\pgfpoint{0cm}{2cm}}
{\pgfpoint{0cm}{2cm}}
{\pgfpoint{3cm}{2cm}}]{\t}}
\end{tikzpicture}
44.5.3 Points on Borders of Objects
The following commands are useful for specifying a point that lies on the border of special shapes. They are
used, for example, by the shape mechanism to determine border points of shapes.
\pgfpointborderrectangle{¸direction point)}{¸corner)}
This command returns a point that lies on the intersection of a line starting at the origin and going
towards the point ¸direction point) and a rectangle whose center is in the origin and whose upper right
corner is at ¸corner).
The ¸direction point) should have length “about 1pt,” but it will be normalized automatically. Never-
theless, the “nearer” the length is to 1pt, the less rounding errors.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (2,1.5);
\pgfpathrectanglecorners{\pgfpoint{-1cm}{-1.25cm}}{\pgfpoint{1cm}{1.25cm}}
\pgfusepath{stroke}
\pgfpathcircle{\pgfpoint{5pt}{5pt}}{2pt}
\pgfpathcircle{\pgfpoint{-10pt}{5pt}}{2pt}
\pgfusepath{fill}
\color{red}
\pgfpathcircle{\pgfpointborderrectangle
{\pgfpoint{5pt}{5pt}}{\pgfpoint{1cm}{1.25cm}}}{2pt}
\pgfpathcircle{\pgfpointborderrectangle
{\pgfpoint{-10pt}{5pt}}{\pgfpoint{1cm}{1.25cm}}}{2pt}
\pgfusepath{fill}
\end{tikzpicture}
\pgfpointborderellipse{¸direction point)}{¸corner)}
This command works like the corresponding command for rectangles, only this time the ¸corner) is the
corner of the bounding rectangle of an ellipse.
293
\begin{tikzpicture}
\draw[help lines] (0,0) grid (2,1.5);
\pgfpathellipse{\pgfpointorigin}{\pgfpoint{1cm}{0cm}}{\pgfpoint{0cm}{1.25cm}}
\pgfusepath{stroke}
\pgfpathcircle{\pgfpoint{5pt}{5pt}}{2pt}
\pgfpathcircle{\pgfpoint{-10pt}{5pt}}{2pt}
\pgfusepath{fill}
\color{red}
\pgfpathcircle{\pgfpointborderellipse
{\pgfpoint{5pt}{5pt}}{\pgfpoint{1cm}{1.25cm}}}{2pt}
\pgfpathcircle{\pgfpointborderellipse
{\pgfpoint{-10pt}{5pt}}{\pgfpoint{1cm}{1.25cm}}}{2pt}
\pgfusepath{fill}
\end{tikzpicture}
44.5.4 Points on the Intersection of Lines
\pgfpointintersectionoflines{¸p)}{¸q)}{¸s)}{¸t)}
This command returns the intersection of a line going through p and q and a line going through s and
t. If the lines do not intersection, an arithmetic overflow will occur.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (2,2);
\draw (.5,0) -- (2,2);
\draw (1,2) -- (2,0);
\pgfpathcircle{%
\pgfpointintersectionoflines
{\pgfpointxy{.5}{0}}{\pgfpointxy{2}{2}}
{\pgfpointxy{1}{2}}{\pgfpointxy{2}{0}}}
{2pt}
\pgfusepath{stroke}
\end{tikzpicture}
44.6 Extracting Coordinates
There are two commands that can be used to “extract” the x- or y-coordinate of a coordinate.
\pgfextractx{¸dimension)}{¸point)}
Sets the T
E
X-¸dimension) to the x-coordinate of the point.
\newdimen\mydim
\pgfextractx{\mydim}{\pgfpoint{2cm}{4pt}}
%% \mydim is now 2cm
\pgfextracty{¸dimension)}{¸point)}
Like \pgfextractx, except for the y-coordinate.
44.7 Internals of How Point Commands Work
As a normal user of pgf you do not need to read this section. It is relevant only if you need to understand
how the point commands work internally.
When a command like \pgfpoint{1cm}{2pt} is called, all that happens is that the two T
E
X-dimension
variables \pgf@x and \pgf@y are set to 1cm and 2pt, respectively. A command like \pgfpathmoveto that
takes a coordinate as parameter will just execute this parameter and then use the values of \pgf@x and
\pgf@y as the coordinates to which it will move the pen on the current path.
since commands like \pgfpointnormalised modify other variables besides \pgf@x and \pgf@y during
the computation of the final values of \pgf@x and \pgf@y, it is a good idea to enclose a call of a command
like \pgfpoint in a T
E
X-scope and then make the changes of \pgf@x and \pgf@y global as in the following
example:
294
...
{ % open scope
\pgfpointnormalised{\pgfpoint{1cm}{1cm}}
\global\pgf@x=\pgf@x % make the change of \pgf@x persist past the scope
\global\pgf@y=\pgf@y % make the change of \pgf@y persist past the scope
}
% \pgf@x and \pgf@y are now set correctly, all other variables are
% unchanged
Since this situation arises very often, the macro \pgf@process can be used to perform the above code:
\pgf@process{¸code)}
Executes the ¸code) in a scope and then makes \pgf@x and \pgf@y global.
Note that this macro is used often internally. For this reason, it is not a good idea to keep anything impor-
tant in the variables \pgf@x and \pgf@y since they will be overwritten and changed frequently. Instead, in-
termediate values can ge stored in the T
E
X-dimensions \pgf@xa, \pgf@xb, \pgf@xc and their y-counterparts
\pgf@ya, \pgf@yb, pgf@yc. For example, here is the code of the command \pgfpointadd:
\def\pgfpointadd#1#2{%
\pgf@process{#1}%
\pgf@xa=\pgf@x%
\pgf@ya=\pgf@y%
\pgf@process{#2}%
\advance\pgf@x by\pgf@xa%
\advance\pgf@y by\pgf@ya}
295
45 Constructing Paths
45.1 Overview
The “basic entity of drawing” in pgf is the path. A path consists of several parts, each of which is either
a closed or open curve. An open curve has a starting point and an end point and, in between, consists of
several segments, each of which is either a straight line or a B´ezier curve. Here is an example of a path (in
red) consisting of two parts, one open, one closed:
start part 1
straight segment
end first segment
end part 1
part 2 (closed)
\begin{tikzpicture}[scale=2]
\draw[thick,red]
(0,0) coordinate (a)
-- coordinate (ab) (1,.5) coordinate (b)
.. coordinate (bc) controls +(up:1cm) and +(left:1cm) .. (3,1) coordinate (c)
(0,1) -- (2,1) -- coordinate (x) (1,2) -- cycle;
\draw (a) node[below] {start part 1}
(ab) node[below right] {straight segment}
(b) node[right] {end first segment}
(c) node[right] {end part 1}
(x) node[above right] {part 2 (closed)};
\end{tikzpicture}
A path, by itself, has no “effect,” that is, it does not leave any marks on the page. It is just a set of
points on the plane. However, you can use a path in different ways. The most natural actions are stroking
(also known as drawing) and filling. Stroking can be imagined as picking up a pen of a certain diameter and
“moving it along the path.” Filling means that everything “inside” the path is filled with a uniform color.
Naturally, the open parts of a path must first be closed before a path can be filled.
In pgf, there are numerous commands for constructing paths, all of which start with \pgfpath. There
are also commands for using paths, though most operations can be performed by calling \pgfusepath with
an appropriate parameter.
As a side-effect, the path construction commands keep track of two bounding boxes. One is the bounding
box for the current path, the other is a bounding box for all paths in the current picture. See Section 45.13
for more details.
Each path construction command extends the current path in some way. The “current path” is a global
entity that persists across T
E
X groups. Thus, between calls to the path construction commands you can
perform arbitrary computations and even open and closed T
E
X groups. The current path only gets “flushed”
when the \pgfusepath command is called (or when the soft-path subsystem is used directly, see Section 61).
45.2 The Move-To Path Operation
The most basic operation is the move-to operation. It must be given at the beginning of paths, though some
path construction command (like \pgfpathrectangle) generate move-tos implicitly. A move-to operation
can also be used to start a new part of a path.
\pgfpathmoveto{¸coordinate)}
This command expects a pgf-coordinate like \pgfpointorigin as its parameter. When the current
path is empty, this operation will start the path at the given ¸coordinate). If a path has already been
partly constructed, this command will end the current part of the path and start a new one.
296
\begin{pgfpicture}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{1cm}{1cm}}
\pgfpathlineto{\pgfpoint{2cm}{1cm}}
\pgfpathlineto{\pgfpoint{3cm}{0.5cm}}
\pgfpathlineto{\pgfpoint{3cm}{0cm}}
\pgfsetfillcolor{examplefill}
\pgfusepath{fill,stroke}
\end{pgfpicture}
\begin{pgfpicture}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{1cm}{1cm}}
\pgfpathlineto{\pgfpoint{2cm}{1cm}}
\pgfpathmoveto{\pgfpoint{2cm}{1cm}} % New part
\pgfpathlineto{\pgfpoint{3cm}{0.5cm}}
\pgfpathlineto{\pgfpoint{3cm}{0cm}}
\pgfsetfillcolor{examplefill}
\pgfusepath{fill,stroke}
\end{pgfpicture}
The command will apply the current coordinate transformation matrix to ¸coordinate) before using it.
The command will update the bounding box of the current path and picture, if necessary.
45.3 The Line-To Path Operation
\pgfpathlineto{¸coordinate)}
This command extends the current path in a straight line to the given ¸coordinate). If this command
is given at the beginning of path without any other path construction command given before (in par-
ticular without a move-to operation), the T
E
X file may compile without an error message, but a viewer
application may display an error message when trying to render the picture.
\begin{pgfpicture}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{1cm}{1cm}}
\pgfpathlineto{\pgfpoint{2cm}{1cm}}
\pgfsetfillcolor{examplefill}
\pgfusepath{fill,stroke}
\end{pgfpicture}
The command will apply the current coordinate transformation matrix to ¸coordinate) before using it.
The command will update the bounding box of the current path and picture, if necessary.
45.4 The Curve-To Path Operation
\pgfpathcurveto{¸support 1)}{¸support 2)}{¸coordinate)}
This command extends the current path with a B´ezier curve from the last point of the path to
¸coordinate). The ¸support 1) and ¸support 2) are the first and second support point of the B´ezier
curve. For more information on B´ezier curve, please consult a standard textbook on computer graphics.
Like the line-to command, this command may not be the first path construction command in a path.
\begin{pgfpicture}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathcurveto
{\pgfpoint{1cm}{1cm}}{\pgfpoint{2cm}{1cm}}{\pgfpoint{3cm}{0cm}}
\pgfsetfillcolor{examplefill}
\pgfusepath{fill,stroke}
\end{pgfpicture}
The command will apply the current coordinate transformation matrix to ¸coordinate) before using it.
The command will update the bounding box of the current path and picture, if necessary. However,
the bounding box is simply made large enough such that it encompasses all of the support points and
the ¸coordinate). This will guarantee that the curve is completely inside the bounding box, but the
bounding box will typically be quite a bit too large. It is not clear (to me) how this can be avoided
without resorting to “some serious math” in order to calculate a precise bounding box.
297
45.5 The Close Path Operation
\pgfpathclose
This command closes the current part of the path by appending a straight line to the start point of the
current part. Note that there is a difference between closing a path and using the line-to operation to
add a straight line to the start of the current path. The difference is demonstrated by the upper corners
of the triangles in the following example:
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfsetlinewidth{5pt}
\pgfpathmoveto{\pgfpoint{1cm}{1cm}}
\pgfpathlineto{\pgfpoint{0cm}{-1cm}}
\pgfpathlineto{\pgfpoint{1cm}{-1cm}}
\pgfpathclose
\pgfpathmoveto{\pgfpoint{2.5cm}{1cm}}
\pgfpathlineto{\pgfpoint{1.5cm}{-1cm}}
\pgfpathlineto{\pgfpoint{2.5cm}{-1cm}}
\pgfpathlineto{\pgfpoint{2.5cm}{1cm}}
\pgfusepath{stroke}
\end{tikzpicture}
45.6 Arc, Ellipse and Circle Path Operations
The path construction commands that we have discussed up to now are sufficient to create all paths that
can be created “at all.” However, it is useful to have special commands to create certain shapes, like circles,
that arise often in practice.
In the following, the commands for adding (parts of) (transformed) circles to a path are described.
\pgfpatharc{¸start angle)}{¸end angle)}{¸radius)/¸y-radius)}
This command appends a part of a circle (or an ellipse) to the current path. Imaging the curve between
¸start angle) and ¸end angle) on a circle of radius ¸radius) (if ¸start angle) < ¸end angle), the curve
goes around the circle counterclockwise, otherwise clockwise). This curve is now moved such that the
point where the curve starts is the previous last point of the path. Note that this command will not
start a new part of the path, which is important for example for filling purposes.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{0cm}{1cm}}
\pgfpatharc{180}{90}{.5cm}
\pgfpathlineto{\pgfpoint{3cm}{1.5cm}}
\pgfpatharc{90}{-45}{.5cm}
\pgfusepath{fill}
\end{tikzpicture}
Saying \pgfpatharc{0}{360}{1cm} “nearly” gives you a full circle. The “nearly” refers to the fact that
the circle will not be closed. You can close it using \pgfpathclose.
If the optional ¸y-radius) is given, the ¸radius) is the x-radius and the ¸y-radius) the y-radius of the
ellipse from which the curve is taken:
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathmoveto{\pgfpointorigin}
\pgfpatharc{180}{45}{2cm/1cm}
\pgfusepath{draw}
\end{tikzpicture}
The axes of the circle or ellipse from which the arc is “taken” always point up and right. However, the
current coordinate transformation matrix will have an effect on the arc. This can be used to, say, rotate
an arc:
298
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgftransformrotate{30}
\pgfpathmoveto{\pgfpointorigin}
\pgfpatharc{180}{45}{2cm/1cm}
\pgfusepath{draw}
\end{tikzpicture}
The command will update the bounding box of the current path and picture, if necessary. Unless
rotation or shearing transformations are applied, the bounding box will be tight.
\pgfpatharcaxes{¸start angle)}{¸end angle)}{¸first axis)}{¸second axis)}
This command is similar to \pgfpatharc. The main difference is how the ellipse or circle is specified
from which the arc is taken. The two parameters ¸first axis) and ¸second axis) are the 0

-axis and the
90

-axis of the ellipse from which the path is taken. Thus, \pgfpatharc{0}{90}{1cm/2cm} has the
same effect as
\pgfpatharcaxes{0}{90}{\pgfpoint{1cm}{0cm}}{\pgfpoint{0cm}{2cm}}
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\draw (0,0) -- (2cm,5mm) (0,0) -- (0cm,1cm);
\pgfpathmoveto{\pgfpoint{2cm}{5mm}}
\pgfpatharcaxes{0}{90}{\pgfpoint{2cm}{5mm}}{\pgfpoint{0cm}{1cm}}
\pgfusepath{draw}
\end{tikzpicture}
\pgfpathellipse{¸center)}{¸first axis)}{¸second axis)}
The effect of this command is to append an ellipse to the current path (if the path is not empty, a
new part is started). The ellipse’s center will be ¸center) and ¸first axis) and ¸second axis) are the
axis vectors. The same effect as this command can also be achieved using an appropriate sequence of
move-to, arc, and close operations, but this command is easier and faster.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathellipse{\pgfpoint{1cm}{0cm}}
{\pgfpoint{1.5cm}{0cm}}
{\pgfpoint{0cm}{1cm}}
\pgfusepath{draw}
\color{red}
\pgfpathellipse{\pgfpoint{1cm}{0cm}}
{\pgfpoint{1cm}{1cm}}
{\pgfpoint{-0.5cm}{0.5cm}}
\pgfusepath{draw}
\end{tikzpicture}
The command will apply coordinate transformations to all coordinates of the ellipse. However, the
coordinate transformations are applied only after the ellipse is “finished conceptually.” Thus, a trans-
formation of 1cm to the right will simply shift the ellipse one centimeter to the right; it will not add
1cm to the x-coordinates of the two axis vectors.
The command will update the bounding box of the current path and picture, if necessary.
\pgfpathcirlce{¸center)}{¸radius)}
A shorthand for \pgfpathellipse applied to ¸center) and the two axis vectors (¸radius), 0) and
(0, ¸radius)).
45.7 Rectangle Path Operations
Another shape that arises frequently is the rectangle. Two commands can be used to add a rectangle to the
current path. Both commands will start a new part of the path.
\pgfpathrectangle{¸corner)}{¸diagonal vector)}
Adds a rectangle to the path whose one corner is ¸corner) and whose opposite corner is given by
¸corner) + ¸diagonal vector).
299
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathrectangle{\pgfpoint{1cm}{0cm}}{\pgfpoint{1.5cm}{1cm}}
\pgfpathrectangle{\pgfpoint{1.5cm}{0.25cm}}{\pgfpoint{1.5cm}{1cm}}
\pgfpathrectangle{\pgfpoint{2cm}{0.5cm}}{\pgfpoint{1.5cm}{1cm}}
\pgfusepath{draw}
\end{tikzpicture}
The command will apply coordinate transformations and update the bounding boxes tightly.
\pgfpathrectanglecorners{¸corner)}{¸opposite corner)}
Adds a rectangle to the path whose two opposing corners are ¸corner) and ¸opposite corner).
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathrectanglecorners{\pgfpoint{1cm}{0cm}}{\pgfpoint{1.5cm}{1cm}}
\pgfusepath{draw}
\end{tikzpicture}
The command will apply coordinate transformations and update the bounding boxes tightly.
45.8 The Grid Path Operation
\pgfpathgrid[¸options)]{¸lower left)}{¸upper right)}
Appends a grid to the current path. That is, a (possibly large) number of parts are added to the path,
each part consisting of a single horizontal or vertical straight line segment.
Conceptually, the origin is part of the grid and the grid is clipped to the rectangle specified by the
¸lower left) and the ¸upper right) corner. However, no clipping occurs (this command just adds parts to
the current path). Rather, the points where the lines enter and leave the “clipping area” are computed
and used to add simple lines to the current path.
Allowed ¸options) are:
• stepx=¸dimension) Sets the horizontal stepping to ¸dimension). Default is 1cm.
• stepy=¸dimension) Sets the vertical stepping to ¸dimension). Default is 1cm.
• step=¸vector) Sets the horizontal stepping to the x-coordinate of ¸vector) and the vertical stepping
to its y-coordinate.
\begin{pgfpicture}
\pgfsetlinewidth{0.8pt}
\pgfpathgrid[step={\pgfpoint{1cm}{1cm}}]
{\pgfpoint{-3mm}{-3mm}}{\pgfpoint{33mm}{23mm}}
\pgfusepath{stroke}
\pgfsetlinewidth{0.4pt}
\pgfpathgrid[stepx=1mm,stepy=1mm]
{\pgfpoint{-1.5mm}{-1.5mm}}{\pgfpoint{31.5mm}{21.5mm}}
\pgfusepath{stroke}
\end{pgfpicture}
The command will apply coordinate transformations and update the bounding boxes tightly. As for
ellipses, the transformations are applied to the “conceptually finished” grid.
\begin{pgfpicture}
\pgftransformrotate{10}
\pgfpathgrid[stepx=1mm,stepy=2mm]{\pgfpoint{0mm}{0mm}}{\pgfpoint{30mm}{30mm}}
\pgfusepath{stroke}
\end{pgfpicture}
300
45.9 The Parabola Path Operation
\pgfpathparabola{¸bend vector)}{¸end vector)}
This command appends two half-parabolas to the current path. The first starts at the current point
and ends at the current point plus ¸bend vector). At his point, it has its bend. The second half parabola
starts at that bend point and end at point that is given by the bend plus ¸end vector).
If you set ¸end vector) to the null vector, you append only a half parabola that goes from the current
point to the bend; by setting ¸bend vector) to the null vector, you append only a half parabola that
goes to current point plus ¸end vector) and has its bend at the current point.
It is not possible to use this command to draw a part of a parabola that does not contain the bend.
\begin{pgfpicture}
% Half-parabola going ‘‘up and right’’
\pgfpathmoveto{\pgfpointorigin}
\pgfpathparabola{\pgfpointorigin}{\pgfpoint{2cm}{4cm}}
\color{red}
\pgfusepath{stroke}
% Half-parabola going ‘‘down and right’’
\pgfpathmoveto{\pgfpointorigin}
\pgfpathparabola{\pgfpoint{-2cm}{4cm}}{\pgfpointorigin}
\color{blue}
\pgfusepath{stroke}
% Full parabola
\pgfpathmoveto{\pgfpoint{-2cm}{2cm}}
\pgfpathparabola{\pgfpoint{1cm}{-1cm}}{\pgfpoint{2cm}{4cm}}
\color{orange}
\pgfusepath{stroke}
\end{pgfpicture}
The command will apply coordinate transformations and update the bounding boxes.
45.10 Sine and Cosine Path Operations
Sine and cosine curves often need to be drawn and the following commands may help with this. However,
they only allow you to append sine and cosine curves in intervals that are multiples of π/2.
\pgfpathsine{¸vector)}
This command appends a sine curve in the interval [0, π/2] to the current path. The sine curve is
squeezed or stretched such that the curve starts at the current point and ends at the current point plus
¸vector).
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,1);
\pgfpathmoveto{\pgfpoint{1cm}{0cm}}
\pgfpathsine{\pgfpoint{1cm}{1cm}}
\pgfusepath{stroke}
\color{red}
\pgfpathmoveto{\pgfpoint{1cm}{0cm}}
\pgfpathsine{\pgfpoint{-2cm}{-2cm}}
\pgfusepath{stroke}
\end{tikzpicture}
The command will apply coordinate transformations and update the bounding boxes.
\pgfpathcosine{¸vector)}
This command appends a cosine curve in the interval [0, π/2] to the current path. The curve is squeezed
or stretched such that the curve starts at the current point and ends at the current point plus ¸vector).
Using several sine and cosine operations in sequence allows you to produce a complete sine or cosine
curve
301
\begin{pgfpicture}
\pgfpathmoveto{\pgfpoint{0cm}{0cm}}
\pgfpathsine{\pgfpoint{1cm}{1cm}}
\pgfpathcosine{\pgfpoint{1cm}{-1cm}}
\pgfpathsine{\pgfpoint{1cm}{-1cm}}
\pgfpathcosine{\pgfpoint{1cm}{1cm}}
\pgfsetfillcolor{examplefill}
\pgfusepath{fill,stroke}
\end{pgfpicture}
The command will apply coordinate transformations and update the bounding boxes.
45.11 Plot Path Operations
There exist several commands for appending plots to a path. These commands are available through the
package pgfbaseplot. They are documented in Section 56.
45.12 Rounded Corners
Normally, when you connect two straight line segments or when you connect two curves that end and start
“at different angles” you get “sharp corners” between the lines or curves. In some cases it is desirable to
produce “rounded corners” instead. Thus, the lines or curves should be shortened a bit and then connected
by arcs.
pgf offers an easy way to achieve this effect, by calling the following two commands.
\pgfsetcornersarced{¸point)}
This command causes all subsequent corners to be replaced by little arcs. The effect of this command
lasts till the end of the current T
E
X scope.
The ¸point) dictates how large the corner arc will be. Consider a corner made by two lines l and r and
assume that the line l comes first on the path. The x-dimension of the ¸point) decides by how much the
line l will be shortened, the y-dimension of ¸point) decides by how much the line r will be shortened.
Then, the shortened lines are connected by an arc.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfsetcornersarced{\pgfpoint{5mm}{5mm}}
\pgfpathrectanglecorners{\pgfpointorigin}{\pgfpoint{3cm}{2cm}}
\pgfusepath{stroke}
\end{tikzpicture}
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfsetcornersarced{\pgfpoint{10mm}{5mm}}
% 10mm entering,
% 5mm leaving.
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{0cm}{2cm}}
\pgfpathlineto{\pgfpoint{3cm}{2cm}}
\pgfpathcurveto
{\pgfpoint{3cm}{0cm}}
{\pgfpoint{2cm}{0cm}}
{\pgfpoint{1cm}{0cm}}
\pgfusepath{stroke}
\end{tikzpicture}
If the x- and y-coordinates of ¸point) are the same and the corner is a right angle, you will get a perfect
quarter circle (well, not quite perfect, but perfect up to six decimals). When the angle is not 90

, you
only get a fair approximation.
More or less “all” corners will be rounded, even the corner generated by a \pgfpathclose command.
(The author is a bit proud of this feature.)
302
\begin{pgfpicture}
\pgfsetcornersarced{\pgfpoint{4pt}{4pt}}
\pgfpathmoveto{\pgfpointpolar{0}{1cm}}
\pgfpathlineto{\pgfpointpolar{72}{1cm}}
\pgfpathlineto{\pgfpointpolar{144}{1cm}}
\pgfpathlineto{\pgfpointpolar{216}{1cm}}
\pgfpathlineto{\pgfpointpolar{288}{1cm}}
\pgfpathclose
\pgfusepath{stroke}
\end{pgfpicture}
To return to normal (unrounded) corners, use \pgfsetcornersarced{\pgfpointorigin}.
Note that the rounding will produce strange and undesirable effects if the lines at the corners are too
short. In this case the shortening may cause the lines to “suddenly extend over the other end” which is
rarely desirable.
45.13 Internal Tracking of Bounding Boxes for Paths and Pictures
The path construction commands keep track of two bounding boxes: One for the current path, which is reset
whenever the path is used and thereby flushed, and a bounding box for the current {pgfpicture}.
The bounding boxes are not accessible by “normal” macros. Rather, two sets of four dimension variables
are used for this, all of which contain the letter @.
\pgf@pathminx
The minimum x-coordinate “mentioned” in the current path. Initially, this is set to 16000pt.
\pgf@pathmaxx
The maximum x-coordinate “mentioned” in the current path. Initially, this is set to −16000pt.
\pgf@pathminy
The minimum y-coordinate “mentioned” in the current path. Initially, this is set to 16000pt.
\pgf@pathmaxy
The maximum y-coordinate “mentioned” in the current path. Initially, this is set to −16000pt.
\pgf@picminx
The minimum x-coordinate “mentioned” in the current picture. Initially, this is set to 16000pt.
\pgf@picmaxx
The maximum x-coordinate “mentioned” in the current picture. Initially, this is set to −16000pt.
\pgf@picminy
The minimum y-coordinate “mentioned” in the current picture. Initially, this is set to 16000pt.
\pgf@picmaxy
The maximum y-coordinate “mentioned” in the current picture. Initially, this is set to −16000pt.
Each time a path construction command is called, the above variables are (globally) updated. To facilitate
this, you can use the following command:
\pgf@protocolsizes{¸x-dimension)}{¸y-dimension)}
Updates all of the above dimension in such a way that the point specified by the two argu-
ments is inside both bounding boxes. For the picture’s bounding box this updating occurs only if
\ifpgf@relevantforpicturesize is true, see below.
For the bounding box of the picture it is not always desirable that every path construction command
affects this bounding box. For example, if you have just used a clip command, you do not want anything
outside the clipping area to affect the bounding box. For this reason, there exists a special “T
E
X if” that
(locally) decides whether updating should be applied to the picture’s bounding box. Clipping will set this if
to false, as will certain other commands.
303
\pgf@relevantforpicturesizefalse
Suppresses updating of the picture’s bounding box.
\pgf@relevantforpicturesizetrue
Causes updating of the picture’s bounding box.
304
46 Snakes
46.1 Overview
A snake is a “way of adding a winding line to a path.” To be a bit more precise, you use snakes to extend the
path and the commands for using snakes start with \pgfpath. However, snakes do not necessarily extend the
path using line-to and curve-to operations; rather, they can also contain move-to operations and, thereby,
cause the path to be split into many subpaths.
As an example, let us consider a simple snake like the zigzag snake. It looks like this:
\begin{tikzpicture}
\draw[snake=zigzag] (0,0) -- (3,0);
\draw[snake=zigzag] (0,-.5) -- (3,-1);
\end{tikzpicture}
The above example demonstrates the two key features of snakes:
1. Snakes are made up from little segments that are repeated several times. For the zigzag snake these
segements look like this: .
2. Snakes “follow along a straight to the target point.” Thus, when the target point is not to the right of
the start point, the snake is rotated appropriately.
In order to use snakes, they first have to be declared. This declaration contains a detailed description of
how each segement of the snake looks like and in what ordering the different possible segments are used.
Once a snake has been declared, it can be used. For this you specify a target point and a snaked line
will be added from the last point current point to the target point.
46.2 Declaring a Snake
46.2.1 Segments
When you declare a snake, you provide a description of how the different segments of the snake will be
rendered. The description of each segment should be given in a way as if the target of the snaked line where
at (+∞, 0) and as if the segment’s start where at the origin. Thus, for example, the segment of the zigzag
might be defined using the following code:
\pgfpathlineto{\pgfpoint{5pt}{5pt}}
\pgfpathlineto{\pgfpoint{15pt}{-5pt}}
\pgfpathlineto{\pgfpoint{20pt}{0pt}}
pgf will ensure that an appropriate coordinate transformation is in place when the snake segment is
added to the path such that the snake segment actually points in the right direction. Also subsequent
snake segments will be transformed such that they are “further along the line” toward the target. All
transformations are setup automatically.
Note that we did not use a \pgfpathmoveto{\pgfpointorigin} at the beginning of the segment code.
Doing so would subdivide the path into numerous subpath. Rather, we assume that the previous segment
caused the current point to be at the origin.
While we can now describe segments, we still need a way to “stop” the snake. Also, pgf needs to know
the width of the different snake segments such that it can translate the next segment correctly. Even though
it would be possible to compute all this information automatically, pgf does not do so and you have to
provide the information “by hand” for each segment.
46.2.2 Snake Automata
Up to now our snakes only have one segment that is repeated again and again. However, we might also like
to have different segments and use rules to describe which segment should be used where. For example, we
might have special segments at the start and at the end.
For snakes we use a mechanism known in computer science as finite automata to describe which segment
is used in each part of the snake. The idea is the following: For the first segment we start in a special state
called the initial state. In this state, and also in all other state later, pgf first computes how much space is
left on the snake. That is, pgf keeps track of the distance to the target. Attached to each state there is a
set of rules of the following form: “If the remaining distance is less than x, switch to state q.” pgf checks
for each of these rules whether it applies and, if so, immediately switches to state q.
305
Only if none of the rules tell us to switch to another state, pgf will execute the state’s code. This code
will add a segment to the path. In addition to the rules there is also width parameter attached to each state.
pgf then translates the coordinate system by this width and reduces the remaining distance. Then, pgf
either stays in the current state or switches to another state, depending on yet another property attached
of the state.
The whole process stops when a special state called final is reached. The segment of this state is added
to the path immediately (it is often empty, though) and the process ends.
46.2.3 The Snake Declaration Command
\pgfdeclaresnake{¸name)}{¸initial state)}{¸states)}
This command declares a new snake called ¸name). The ¸states) argument contains a description of the
snake automaton’s states and the transitions between them. The ¸initial state) is the state in which the
automaton starts.
The ¸states) argument should consist of \state commands, one for each state of the snake automaton.
The \state command is defined only when the ¸states) argument is executed.
\pgfdeclaresnake{example snake}{initial}
{
\state{initial}[width=10pt]
{
\pgfpathlineto{\pgfpoint{0pt}{5pt}}
\pgfpathlineto{\pgfpoint{5pt}{5pt}}
\pgfpathlineto{\pgfpoint{5pt}{-5pt}}
\pgfpathlineto{\pgfpoint{10pt}{-5pt}}
\pgfpathlineto{\pgfpoint{10pt}{0pt}}
}
\state{final}
{
\pgfpathlineto{\pgfpoint{\pgfsnakeremainingdistance}{0pt}}
}
}
\tikz{\draw[snake=example snake] (0,0) -- (3,0);
\draw[snake=example snake] (0,-1) -- (3,-3);}
The \state command works as follows:
\state{¸name)}[¸options)]{¸code)}
This command declares a new state inside the current snake automaton. The state is names ¸name).
When pgf is in state ¸name), the following things happen:
1. The ¸options) are parsed. This may lead, see below, to a state switch. When this happens, the
following steps are not executed. The ¸options) are executed one after the other in the given
order. If an option causes a state switch, the switch is immediate, even if later options might
cause a different state switch.
2. The ¸code) is executed. When this happens, the coordinate transformation will be setup such
that the origin is at the start of the segment and such that the target lies in the direction
“east.”
3. After the ¸code) has been executed, the state switches to whatever state has been specified
inside the ¸options) using the next state option. If no next state has been specified, the
state stays the same.
The following options are allowed inside the ¸options):
• switch if less than=¸dimension) to ¸new state) When this option is encountered, pgf
checks whether the remaining distance to the target is less than ¸dimension). If so, an imme-
diate state switch to ¸new state) occurs.
• width=¸dimension) This option tells pgf the width of the segment if it is appended to the
path. The coordinate system will, afterward, be translated by ¸dimension) to ensure that the
next segment starts where the current one ended.
In addition, this option will cause an immediate switch to the state final if the re-
maining distance is less than ¸dimension). The effect is the same as if you had said
switch if less than=¸dimension) to final just before the width option.
306
• next state=¸new state) After the segment has been added to the path, a state switch to ¸new
state) is performed. If this option is not given, the next state is the same as the current state.
There are two dimensions that are useful inside snake automata:
\pgfsnakeremainingdistance
This T
E
X dimension holds the remaining distance to the target.
\pgfsnakecompleteddistance
This T
E
X dimension holds the distance already completed on the snake.
As a final example we present a more complicated snake that makes use of the different options:
\pgfdeclaresnake{complicated example snake}{initial}
{
\state{initial}[width=5pt,next state=up]
{ \pgfpathlineto{\pgfpoint{5pt}{0pt}} }
\state{up}[width=5pt,next state=down]
{
\ifdim\pgfsnakeremainingdistance>\pgfsnakecompleteddistance
% Growing
\pgfpathlineto{\pgfpoint{0pt}{\pgfsnakecompleteddistance}}
\pgfpathlineto{\pgfpoint{5pt}{\pgfsnakecompleteddistance}}
\pgfpathlineto{\pgfpoint{5pt}{0pt}}
\else
% Shrinking
\pgfpathlineto{\pgfpoint{0pt}{\pgfsnakeremainingdistance}}
\pgfpathlineto{\pgfpoint{5pt}{\pgfsnakeremainingdistance}}
\pgfpathlineto{\pgfpoint{5pt}{0pt}}
\fi%
}
\state{down}[width=5pt,next state=up]
{
\ifdim\pgfsnakeremainingdistance>\pgfsnakecompleteddistance
% Growing
\pgfpathlineto{\pgfpoint{0pt}{-\pgfsnakecompleteddistance}}
\pgfpathlineto{\pgfpoint{5pt}{-\pgfsnakecompleteddistance}}
\pgfpathlineto{\pgfpoint{5pt}{0pt}}
\else
% Shrinking
\pgfpathlineto{\pgfpoint{0pt}{-\pgfsnakeremainingdistance}}
\pgfpathlineto{\pgfpoint{5pt}{-\pgfsnakeremainingdistance}}
\pgfpathlineto{\pgfpoint{5pt}{0pt}}
\fi%
}
\state{final}
{
\pgfpathlineto{\pgfpoint{\pgfsnakeremainingdistance}{0pt}}
}
}
\begin{tikzpicture}[snake=complicated example snake]
\draw[snake] (0,0) -- (3,0);
\fill[red!50,snake,rounded corners=2pt] (.5,-2) -- ++(2.5,-2.5)
[snake=none] -- ++(0,-.5) -| (0,-2) -- cycle;
\end{tikzpicture}
46.2.4 Predefined Snakes
Only two very simple and basic snakes are predefined when you load pgf. For more interesting snakes you
can use the package pgflibrarysnakes.
Snake lineto
This most trivial of all snakes is simply a straight line. This snake is typically not used in a “stand
alone” fashion. Rather, it is useful as a “subsnake” of the \pgfsnakesto command.
Snake moveto
Arguably, this snake is even simpler than the previous snake. This snake consists of a simple move-to
operation. Like the lineto snake, it is useful as a “subsnake” of the \pgfsnakesto command.
307
46.3 Using Snakes
Three commands can be used to use a snake.
\pgfpathsnakealongvector{¸snake)}{¸length)}{¸vector)}
This command will append the ¸snake) to the current path. The length of the snake is given by ¸length).
The ¸vector) should be a normalizes vector (a vector having length 1pt) pointing in the direction in
which the snake should grow.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathmoveto{\pgfpoint{1cm}{1cm}}
\pgfpathsnakealongvector{zigzag}{2cm}{\pgfpointpolar{30}{1pt}}
\pgfusepath{stroke}
\end{tikzpicture}
\pgfpathsnakesto{¸snake list)}{¸target)}
This command will append the snakes in the ¸snake list) to the current path such that it ends at ¸point).
This command calls the previous one (repeatedly, possibly) after having computed the distance from
the current point to ¸target) and normalized the vector connecting the current point to the target.
The ¸snake list) is a comma-separated list of pairs consisting of a snake name in curly braces and a
distance in curly braces. For example, a snake list might be {lineto}{1cm},{moveto}{2cm}. This is
interpreted as follows: Use a lineto snake for the first centimeter of the way to the ¸target), then use
the moveto snake for the next two centimeters.
No attempt is made to check or ensure that the distances of the “subsnakes’ add up to the distnace
from the current point to the ¸target). However, it is possible to refer to this distance inside the ¸snake
list): While the snake list is processed, the two T
E
X-dimensions \pgfsnakeremainingdistance and
\pgfsnakecompleteddistance will be set to the correct values. For example, consider the example
from a above, consisting of a lineto and a moveto snake. Suppose the distance to ¸target) where 5cm.
Then, when the length of the first subsnake is computed, the remaining distance willbe set to 5cm and
the completed distance to 0cm. When the length of the second subsnake is computed, the remaining
distance wil be 4cm and the completed distance will be 1cm.
Here are some useful examles: {zigzag}{\pgfsnakeremainingdistance} is a ¸snake list) that consists
only of a zigzag snake whose length is the total distance from the current point to the ¸target). Next,
to get a snake that is a zigzag snake that is preceeded and succeeded by 3mm of lineto, you can use
the following:
\begin{tikzpicture}
\draw[help lines](0,0) grid (3,2);
\pgfpathmoveto{\pgfpointorigin}
\pgfpathsnakesto{%
{lineto}{3mm},%
{zigzag}{\pgfsnakeremainingdistance-6mm},%
{lineto}{3mm}}
{\pgfpoint{3cm}{2cm}}
\pgfusepath{stroke}
\end{tikzpicture}
Note that the computation of the distance may be imprecise. In general, the placement precision of the
snakes will not be perfect.
\pgfpathsnaketo{¸snake)}{¸target)}
This command is just a shortcut for calling \pgfpathsnakesto for a single snake whose length is
\pgfsnakeremainingdistance.
\begin{tikzpicture}
\draw[help lines] (0,0) grid (3,2);
\pgfpathmoveto{\pgfpoint{1cm}{1cm}}
\pgfpathsnaketo{zigzag}{\pgfpoint{3cm}{2cm}}
\pgfusepath{stroke}
\end{tikzpicture}
308
As was already mentioned, when each segment of the snake is added to the path, an appropriate coordinate
transformation will be in force. It is sometimes useful to add an additional transformation locally. For
example, by reflecting everything around the x-axis right before each segment is added, the snake will
effectively be mirrored along the path. The following command allows you to install such a “last minute
transformation.”
\pgfsetsnakesegmenttransformation{¸code)}
The ¸code) will be executed at the very beginning of each segment. Normally, this be a transformation
command that changes the y-axis in some way.
309
47 Using Paths
47.1 Overview
Once a path has been constructed, it can be used in different ways. For example, you can draw the path or
fill it or use it for clipping.
Numerous graph parameters influence how a path will be rendered. For example, when you draw a path,
the line width is important as well as the dashing pattern. The options that govern how paths are rendered
can all be set with commands starting with \pgfset. All options that influence how a path is rendered
always influence the complete path. Thus, it is not possible to draw part of a path using, say, a red color
and drawing another part using a green color. To achieve such an effect, you must use two paths.
In detail, paths can be used in the following ways:
1. You can stroke (also known as draw) a path.
2. You can fill a path with a uniform color.
3. You can clip subsequent renderings against the path.
4. You can shade a path.
5. You can use the path as bounding box for the whole picture.
You can also perform any combination of the above, though it makes no sense to fill and shade a path at
the same time.
To perform (a combination of) the first three actions, you can use the following command:
\pgfusepath{¸actions)}
Applies the given ¸actions) to the current path. Afterwards, the current path is (globally) empty. The
following actions are possible:
• fill fills the path. See Section 47.3 for further details.
\begin{pgfpicture}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{1cm}{1cm}}
\pgfpathlineto{\pgfpoint{1cm}{0cm}}
\pgfusepath{fill}
\end{pgfpicture}
• stroke strokes the path. See Section 47.2 for further details.
\begin{pgfpicture}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{1cm}{1cm}}
\pgfpathlineto{\pgfpoint{1cm}{0cm}}
\pgfusepath{stroke}
\end{pgfpicture}
• clip clips all subsequent drawings against the path. See Section 47.4 for further details.
\begin{pgfpicture}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{1cm}{1cm}}
\pgfpathlineto{\pgfpoint{1cm}{0cm}}
\pgfusepath{stroke,clip}
\pgfpathcircle{\pgfpoint{1cm}{1cm}}{0.5cm}
\pgfusepath{fill}
\end{pgfpicture}
• discard discards the path, that is, it is not used at all. Giving this option (alone) has the same
effect as giving an empty options list.
When more than one of the first three actions are given, they are applied in the above ordering, regardless
of their ordering in ¸actions). Thus, {stroke,fill} and {fill,stroke} have the same effect.
To shade a path, use the \pgfshadepath command, which is explained in Section 55.
310
47.2 Stroking a Path
When you use \pgfusepath{stroke} to stroke a path, several graphic parameters influence how the path
is drawn. The commands for setting these parameters are explained in the following.
Note that all graphic parameters apply to the path as a whole, never only to a part of it.
All graphic parameters are local to the current {pgfscope}, but they persists past T
E
X groups, except
for the interior rule (even-odd or nonzero) and the arrow tip kinds. The latter graphic parameters only
persist till the end of the current T
E
X group, but this may change in the future, so do not count on this.
47.2.1 Graphic Parameter: Line Width
\pgfsetlinewidth{¸line width)}
This command sets the line width for subsequent strokes (in the current pgfscope). The line width is
given as a normal T
E
X dimension like 0.4pt or 1mm.
\begin{pgfpicture}
\pgfsetlinewidth{1mm}
\pgfpathmoveto{\pgfpoint{0mm}{0mm}}
\pgfpathlineto{\pgfpoint{2cm}{0mm}}
\pgfusepath{stroke}
\pgfsetlinewidth{2\pgflinewidth} % double in size
\pgfpathmoveto{\pgfpoint{0mm}{5mm}}
\pgfpathlineto{\pgfpoint{2cm}{5mm}}
\pgfusepath{stroke}
\end{pgfpicture}
\pgflinewidth
You can access the current line width via the T
E
X dimension \pgflinewidth. It will be set to the correct
line width, that is, even when a T
E
X group closed, the value will be correct since it is set globally, but
when a {pgfscope} closes, the value is set to the correct value it had before the scope.
47.2.2 Graphic Parameter: Caps and Joins
\pgfsetbuttcap
Sets the line cap to a butt cap. See Section 12.2.1 for an explanation of what this is.
\pgfsetroundcap
Sets the line cap to a round cap. See again Section 12.2.1.
\pgfsetrectcap
Sets the line cap to a square cap. See again Section 12.2.1.
\pgfsetroundjoin
Sets the line join to a round join. See again Section 12.2.1.
\pgfsetbeveljoin
Sets the line join to a bevel join. See again Section 12.2.1.
\pgfsetmiterjoin
Sets the line join to a miter join. See again Section 12.2.1.
\pgfsetmiterlimit{¸miter limit factor)}
Sets the miter limit to ¸miter limit factor). See again Section 12.2.1.
47.2.3 Graphic Parameter: Dashing
\pgfsetdash{¸list of even length of dimensions)}{¸phase)}
Sets the dashing of a line. The first entry in the list specifies the length of the first solid part of the list.
The second entry specifies the length of the following gap. Then comes the length of the second solid
part, following by the length of the second gap, and so on. The ¸phase) specifies where the first solid
part starts relative to the beginning of the line.
311
\begin{pgfpicture}
\pgfsetdash{{0.5cm}{0.5cm}{0.1cm}{0.2cm}}{0cm}
\pgfpathmoveto{\pgfpoint{0mm}{0mm}}
\pgfpathlineto{\pgfpoint{2cm}{0mm}}
\pgfusepath{stroke}
\pgfsetdash{{0.5cm}{0.5cm}{0.1cm}{0.2cm}}{0.1cm}
\pgfpathmoveto{\pgfpoint{0mm}{1mm}}
\pgfpathlineto{\pgfpoint{2cm}{1mm}}
\pgfusepath{stroke}
\pgfsetdash{{0.5cm}{0.5cm}{0.1cm}{0.2cm}}{0.2cm}
\pgfpathmoveto{\pgfpoint{0mm}{2mm}}
\pgfpathlineto{\pgfpoint{2cm}{2mm}}
\pgfusepath{stroke}
\end{pgfpicture}
Use \pgfsetdash{}{0pt} to get a solid dashing.
47.2.4 Graphic Parameter: Stroke Color
\pgfsetstrokecolor{¸color)}
Sets the color used for stroking lines to ¸color), where ¸color) is a L
A
T
E
X color like red or black!20!red.
Unlike the \color command, the effect of this command lasts till the end of the current {pgfscope}
and not till the end of the current T
E
X group.
The color used for stroking may be different from the color used for filling. However, a \color command
will always “immediately override” any special settings for the stroke and fill colors.
In plain T
E
X, this command will also work, but the problem of defining a color arises. After all, plain
T
E
X does not provide L
A
T
E
X colors. For this reason, pgf implements a minimalistic “emulation” of the
\definecolor, \colorlet, and \color commands. Only gray-scale and rgb colors are supported. For
most cases this turns out to be enough.
\begin{pgfpicture}
\pgfsetlinewidth{1pt}
\color{red}
\pgfpathcircle{\pgfpoint{0cm}{0cm}}{3mm} \pgfusepath{fill,stroke}
\pgfsetstrokecolor{black}
\pgfpathcircle{\pgfpoint{1cm}{0cm}}{3mm} \pgfusepath{fill,stroke}
\color{red}
\pgfpathcircle{\pgfpoint{2cm}{0cm}}{3mm} \pgfusepath{fill,stroke}
\end{pgfpicture}
\pgfsetcolor{¸color)}
Sets both the stroke and fill color. The difference to the normal \color command is that the effect lasts
till the end of the current {pgfscope}, not only till the end of the current T
E
X group.
47.2.5 Graphic Parameter: Stroke Opacity
\pgfsetstrokeopacity{¸value)}
Sets the opacity of stroking operations. The ¸value) should be a number between 0 and 1, where 1
means “fully opaque” and 0 means “fully transparent.” A value like 0.5 will cause paths to be stroked
in a semitransparent way.
Note: For PostScript output, opacity is rendered correctly only with the most recent versions of
GhostScript. Printers and other programs will ignore the opacity setting.
\begin{pgfpicture}
\pgfsetlinewidth{5mm}
\color{red}
\pgfpathcircle{\pgfpoint{0cm}{0cm}}{10mm} \pgfusepath{stroke}
\color{black}
\pgfsetstrokeopacity{0.5}
\pgfpathcircle{\pgfpoint{1cm}{0cm}}{10mm} \pgfusepath{stroke}
\end{pgfpicture}
312
47.2.6 Graphic Parameter: Arrows
After a path has been drawn, pgf can add arrow tips at the ends. Currently, it will only add arrows correctly
at the end of paths that consist of a single open part. For other paths, like closed paths or path consisting
of multiple parts, the result is not defined.
\pgfsetarrowsstart{¸arrow kind)}
Sets the arrow tip kind used at the start of a (possibly curved) path. When this option is used, the line
will often be slightly shortened to ensure that the tip of the arrow will exactly “touch” the “real” start
of the line.
To “clear” the start arrow, say \pgfsetarrowsstart{}.
\begin{pgfpicture}
\pgfsetarrowsstart{latex}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{1cm}{0cm}}
\pgfusepath{stroke}
\pgfsetarrowsstart{to}
\pgfpathmoveto{\pgfpoint{0cm}{2mm}}
\pgfpathlineto{\pgfpoint{1cm}{2mm}}
\pgfusepath{stroke}
\end{pgfpicture}
The effect of this command persists only till the end of the current T
E
X scope.
The different possible arrow kinds are explained in Section 48.
\pgfsetarrowsend{¸arrow kind)}
Sets the arrow tip kind used at the end of a path.
\begin{pgfpicture}
\pgfsetarrowsstart{latex}
\pgfsetarrowsend{to}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{1cm}{0cm}}
\pgfusepath{stroke}
\end{pgfpicture}
\pgfsetarrows{¸start kind)-¸end kind)}
Sets the start arrow kind to ¸start kind) and the end kind to ¸end kind).
\begin{pgfpicture}
\pgfsetarrows{latex-to}
\pgfpathmoveto{\pgfpointorigin}
\pgfpathlineto{\pgfpoint{1cm}{0cm}}
\pgfusepath{stroke}
\end{pgfpicture}
\pgfsetshortenstart{¸dimension)}
This command will shortened the start of every stroked path by the given dimension. This shortening
is done in addition to automatic shortening done by a start arrow, but it can be used even if no start
arrow is given.
This command is useful if you wish arrows or lines to “stop shortly before” a given point.
\begin{pgfpicture}
\pgfpathcircle{\pgfpointorigin}{5mm}
\pgfusepath{stroke}
\pgfsetarrows{latex-}
\pgfsetshortenstart{4pt}
\pgfpathmoveto{\pgfpoint{5mm}{0cm}} % would be on the circle
\pgfpathlineto{\pgfpoint{2cm}{0cm}}
\pgfusepath{stroke}
\end{pgfpicture}
\pgfsetshortenend{¸dimension)}
Works like \pgfsetshortenstart.
313
47.3 Filling a Path
Filling a path means coloring every interior point of the path with the current fill color. It is not always
obvious whether a point is “inside” a path when the path is self-intersecting and/or consists or multiple
parts. In this case either the nonzero winding number rule or the even-odd crossing number rule is used to
decide, which points lie “inside.” These rules are explained in Section 12.3.
47.3.1 Graphic Parameter: Interior Rule
You can set which rule is used using the following commands:
\pgfseteorule
Dictates that the even-odd rule is used in subsequent fillings in the current T
E
X scope. Thus, for once,
the effect of this command does not persist past the current T
E
X scope.
\begin{pgfpicture}
\pgfseteorule
\pgfpathcircle{\pgfpoint{0mm}{0cm}}{7mm}
\pgfpathcircle{\pgfpoint{5mm}{0cm}}{7mm}
\pgfusepath{fill}
\end{pgfpicture}
\pgfsetnonzerorule
Dictates that the nonzero winding number rule is used in subsequent fillings in the current T
E
X scope.
This is the default.
\begin{pgfpicture}
\pgfsetnonzerorule
\pgfpathcircle{\pgfpoint{0mm}{0cm}}{7mm}
\pgfpathcircle{\pgfpoint{5mm}{0cm}}{7mm}
\pgfusepath{fill}
\end{pgfpicture}
47.3.2 Graphic Parameter: Filling Color
\pgfsetfillcolor{¸color)}
Sets the color used for filling paths to ¸color). Like the stroke color, the effect lasts only till the next
use of \color.
47.3.3 Graphic Parameter: Fill Opacity
\pgfsetfillopacity{¸value)}
Sets the opacity of filling operations. As for stroking, the ¸value) should be a number between 0 and 1.
The “filling transparency” will also be used for text and images.
\begin{tikzpicture}
\pgfsetfillopacity{0.5}
\fill[red] (90:1cm) circle (11mm);
\fill[green] (210:1cm) circle (11mm);
\fill[blue] (-30:1cm) circle (11mm);
\end{tikzpicture}
47.4 Clipping a Path
When you add the clip option, the current path is used for clipping subsequent drawings. The same rule
as for filling is used to decide whether a point is inside or outside the path, that is, either the even-odd rule
or the nonzero rule.
314
Clipping never enlarges the clipping area. Thus, when you clip against a certain path and then clip again
against another path, you clip against the intersection of both.
The only way to enlarge the clipping path is to end the {pgfscope} in which the clipping was done. At
the end of a {pgfscope} the clipping path that was in force at the beginning of the scope is reinstalled.
47.5 Using a Path as a Bounding Box
When you add the use as bounding box option, the bounding box of the picture will be enlarged such
that the path in encompassed, but any subsequent paths of the current T
E
X scope will not have any effect
on the size of the bounding box. Typically, you use this command at the very beginning of a {pgfpicture}
environment.
Left right.
Left
\begin{pgfpicture}
\pgfpathrectangle{\pgfpointorigin}{\pgfpoint{2ex}{1ex}}
\pgfusepath{use as bounding box} % draws nothing
\pgfpathcircle{\pgfpointorigin}{2ex}
\pgfusepath{stroke}
\end{pgfpicture}
right.
315
48 Arrow Tips
48.1 Overview
48.1.1 When Does PGF Draw Arrow Tips?
pgf offers an interface for placing arrow tips at the end of lines. The interface works as follows:
1. You (or someone else) assigns a name to a certain kind of arrow tips. For example, the arrow tip latex
is the arrow tip used by the standard L
A
T
E
X picture environment; the arrow tip to looks like the tip of
the arrow in T
E
X’s \to command; and so on.
This is done once at the beginning of the document.
2. Inside some picture, at some point you specify that in the current scope from now on you would like
tips of, say, kind to to be added at the end and/or beginning of all paths.
When an arrow kind has been installed and when pgf is about to stroke a path, the following things
happen:
(a) The beginning and/or end of the path is shortened appropriately.
(b) The path is stroked.
(c) The arrow tip is drawn at the beginning and/or end of the path, appropriately rotated and
appropriately resized.
In the above description, there are a number of “appropriately.” The exact details are not quite trivial
and described later on.
48.1.2 Meta-Arrow Tips
In pgf, arrows are “meta-arrows” in the same way that fonts in T
E
X are “meta-fonts.” When a meta-arrow
is resized, it is not simply scaled, but a possibly complicated transformation is applied to the size.
A meta-font is not one particular font at a specific size with a specific stroke width (and with a large
number of other parameters being fixed). Rather, it is a “blueprint” (actually, more like a program) for
generating such a font at a particular size and width. This allows the designer of a meta-font to make sure
that, say, the font is somewhat thicker and wider at very small sizes. To appreciate the difference: Compare
the following texts: “Berlin” and “Berlin”. The first is a “normal” text, the second is the tiny version
scaled by a factor of two. Obviously, the first look better. Now, compare “Berlin” and “Berlin”. This time, the
normal text was scaled down, while the second text is a “normal” tiny text. The second text is easier to
read.
pgf’s meta-arrows work in a similar fashion: The shape of an arrow tip can vary according to the line
width of the arrow tip is used. Thus, an arrow tip drawn at a line width of 5pt will typically not be five
times as large as an arrow tip of line width 1pt. Instead, the size of the arrow will get bigger only slowly as
the line width increases.
To appreciate the difference, here are the latex and to arrows, as drawn by pgf at four different sizes:
line width is 0.1pt
line width is 0.4pt
line width is 1.2pt
line width is 5pt
line width is 0.1pt
line width is 0.4pt
line width is 1.2pt
line width is 5pt
Here, by comparison, is the same arrow when it is simply “resized” (as done by most programs):
line width is 0.1pt
line width is 0.4pt
line width is 1.2pt
line width is 5pt
line width is 0.1pt
line width is 0.4pt
line width is 1.2pt
line width is 5pt
As can be seen, simple scaling produces arrow tips that are way too large at larger sizes and way too
small at smaller sizes.
316
48.2 Declaring an Arrow Tip Kind
To declare an arrow kind “from scratch,” the following command is used:
\pgfarrowsdeclare{¸start name)}{¸end name)}{¸extend code)}{¸arrow tip code)}
This command declares a new arrow kind. An arrow kind has two names, which will typically be the
same. When the arrow tip needs to be drawn, the ¸arrow tip code) will be invoked, but the canvas
transformation is setup beforehand to a rotation such that when an arrow tip pointing right is specified,
the arrow tip that is actually drawn points in the direction of the line.
Naming the arrow kind. The ¸start name) is the name used for the arrow tip when it is at the start
of a path, the ¸end name) is the name used at the end of a path. For example, the arrow kind that looks
like a parenthesis has the ¸start name) ( and the ¸end name) ) so that you can say \pgfsetarrows{(-)}
to specify that you want parenthesis arrows and both ends.
The ¸end name) and ¸start name) can be quite arbitrary and may contain spaces.
Basics of the arrow tip code. Let us next have a look at the ¸arrow tip code). This code will be
used to draw the arrow tip when pgf thinks this is necessary. The code should draw an arrow that
“points right,” which means that is should draw an arrow at the end of a line coming from the left and
ending at the origin.
As an example, suppose we wanted to declare an arrow tip consisting of two arcs, that is, we want the
arrow tip to look more or less like the red part of the following picture:
\begin{tikzpicture}[line width=3pt]
\draw (-2,0) -- (0,0);
\draw[red,join=round,cap=round]
(-10pt,10pt) arc (180:270:10pt) arc (90:180:10pt);
\end{tikzpicture}
We could use the following as ¸arrow tip code) for this:
\pgfarrowsdeclare